Machine learning for identifying Randomized Controlled Trials: An evaluation and practitioner's guide

Artificial intelligence in key pricing, reimbursement, and market access (PRMA) processes: better, faster, cheaper-can you really pick two?

The rapid evolution of large language models (LLMs) and machine learning (ML) presents both significant opportunities and challenges for market access processes. These sophisticated AI systems, built on transformer architectures and extensive datasets, offer potential to forecast claims and decisions of health technology assessment (HTA) agencies and streamline processes, such as systematic literature reviews and HTA submissions. Furthermore, the analysis of real-world data-also for deriving causal relationships-is being discussed intensively. Despite notable advancements, their adoption in key PRMA processes is still limited at present, with only a small fraction of submissions to HTA bodies incorporating AI. Key barriers include stringent transparency requirements, the necessity of explainability and human oversight in data analyses, and the highly sensitive nature of text drafting-especially in cases where reimbursement decisions or pricing negotiations balance on a knife's edge. These requirements are often not met due to the immaturity of many AI applications, which still lack the necessary precision, reliability, and contextual understanding. Moreover, AI-generated evidence has yet to prove its validity before it can supplement or replace traditional study designs, such as randomized controlled trials (RCTs), which are critical for HTA decisions. Additionally, the environmental and financial costs of training LLMs require careful assessment. This paper explores various current AI applications, their limitations, and future prospects in key PRMA processes from a German perspective while also considering the broader implications of the EU Health Technology Assessment Regulation (HTAR). It concludes that while AI hold transformative potential, its integration into workflows must be approached cautiously, with incremental adoption, and close collaboration between industry, HTA agencies, and academia. Demonstrating robust, unbiased comparative evidence-showcasing superior performance and cost savings over traditional methods-could accelerate the adoption process.

Clinical utility of limited channel sleep studies versus polysomnography for obstructive sleep apnoea

BACKGROUND

Obstructive sleep apnoea (OSA) is a common cause of sleep disturbance, characterised by the presence of repetitive upper airway obstruction during sleep. OSA is associated with sleepiness during the day, reduced quality of life and an increased risk of cardiovascular disease. OSA can be diagnosed using several different strategies. The current reference test is fully supervised polysomnography, which is expensive and time-consuming. Other diagnostic tests, referred to as limited channel sleep studies because they include fewer parameters than polysomnography, are less resource-intensive but may also have different diagnostic performances, resulting in a difference in clinical outcomes.

OBJECTIVES

To assess the clinical impact (outcome on a participant level) of a strategy where treatment follows diagnostic testing (test-treatment combination) using limited channel sleep studies compared to polysomnography in people with suspected obstructive sleep apnoea (OSA).

SEARCH METHODS

We searched two databases (CENTRAL, MEDLINE) up to 11 May 2023 using search terms related to OSA and polysomnography developed by our information specialist.

SELECTION CRITERIA

We included randomised controlled trials that compared any limited channel sleep studies with Level I fully supervised polysomnography in adults (aged 18 years and older) with suspected OSA. Our primary outcome was sleepiness, and our secondary outcomes were quality of life, all-cause mortality, cardiovascular events and correlating risk factors, continuous positive airway pressure (CPAP) usage, serious adverse events, and cost-effectiveness.

DATA COLLECTION AND ANALYSIS

Four review authors extracted data from the included trials and assessed the risk of bias. We summarised treatment effects using random-effects meta-analyses and expressed as mean difference (MD) or standardised mean difference (SMD) with corresponding 95% confidence intervals (CI) where possible. We used GRADE to assess the certainty of the evidence.

MAIN RESULTS

We included three trials with 1143 participants. One trial compared Level III sleep studies to a Level I fully supervised polysomnography, one trial compared Level IV sleep studies to Level I sleep studies, and one trial compared Level IV sleep studies versus Level III sleep studies versus Level I sleep studies. The follow-up of these trials ranged from four to six months. Level III sleep studies versus Level I sleep studies There is high-certainty evidence that Level III sleep studies result in little to no difference in sleepiness (MD 0.47, 95% CI -0.23 to 1.18; P = 0.19, I2 = 0%; 2 trials, 701 participants) or quality of life (SMD 0.01, 95% CI -0.14 to 0.16; P = 0.93, I2 = 0%; 2 trials, 701 participants) compared to Level I sleep studies. Level III sleep studies are also probably slightly more cost-effective (moderate-certainty evidence). There is low-certainty evidence that they may result in little to no difference in cardiovascular events and correlating risk factors, CPAP adherence (MD -0.18 hours per day, 95% CI -0.56 to 0.20; P = 0.36, I2 = 0%; 2 trials, 360 participants) or serious adverse events. Level IV sleep studies versus Level I sleep studies There is low-certainty evidence that Level IV sleep studies may not increase sleepiness compared to Level I sleep studies (MD 0.66, 95% CI -0.41 to 1.72; P = 0.23, I2 = 39%; 2 trials, 573 participants). Additionally, there is low-certainty evidence that they may result in little to no difference in cardiovascular events and correlating risk factors. For quality of life, CPAP adherence, serious adverse events and cost-effectiveness, the evidence is very uncertain. None of the included trials reported on all-cause mortality.

AUTHORS' CONCLUSIONS

Level III sleep studies may result in little to no difference in clinical outcomes when compared to Level 1 sleep studies in people with suspected OSA. Level IV sleep studies may not increase sleepiness and may result in little to no difference in cardiovascular events and correlating risk factors compared to Level I sleep studies; the evidence was too uncertain to make statements for other outcomes. Overall, the body of evidence was limited, therefore more trials making this comparison are necessary, as are trials with a longer follow-up duration.

Psychological therapies for post-traumatic stress disorder in children and adolescents

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the effects (i.e. benefits and harms) of psychological therapies for post-traumatic stress disorder and complex post-traumatic stress disorder in children and adolescents.

Lutein and zeaxanthin for reducing morbidity and mortality in preterm infants

BACKGROUND

Lutein and zeaxanthin are nutrients with antioxidant properties found in the macula of the eye and brain tissue. They have been reported to play a role in reducing oxidative damage, especially in the eyes and possibly in other organ systems. Oxygen free radicals are one of the agents postulated to cause tissue damage in preterm infants, which leads to morbidities such as retinopathy of prematurity (ROP), intraventricular haemorrhage (IVH), and necrotising enterocolitis (NEC). Supplementation with lutein and zeaxanthin may reduce oxidative damage, hence reducing morbidity and mortality in preterm infants.

OBJECTIVES

To assess the effectiveness of lutein and zeaxanthin supplementation in reducing morbidity and mortality in preterm infants.

SEARCH METHODS

We conducted searches up to 17 December 2024 in CENTRAL, MEDLINE, Embase, and two trial registries. We also searched the reference lists of included studies, and related reviews and studies.

SELECTION CRITERIA

We included randomised controlled trials (RCTs), cluster-RCT, cross-over trials, and quasi-RCTs that compared lutein and zeaxanthin supplementation against placebo or no supplementation for preterm infants less than 37 completed weeks' postmenstrual age.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods. Our primary outcomes were the incidence of any stage of ROP, incidence of ROP stage 3 and above, incidence of visual impairment, and mortality assessed throughout the neonatal intensive care unit (NICU) stay. Secondary outcomes included the incidence of IVH, incidence of NEC, and any reported adverse effects. We used GRADE to assess the certainty of the evidence.

MAIN RESULTS

We included five studies (666 preterm infants) that compared lutein and zeaxanthin supplementation versus control (placebo or no supplementation). All five studies were conducted in high-income countries (Italy and the USA). We did not find any studies comparing lutein or zeaxanthin separately versus placebo or no supplementation. Most of the studies had a low risk of bias in most key domains, such as allocation concealment and blinding. The evidence suggests that lutein and zeaxanthin supplementation probably has little or no effect on ROP (any stage) when comparing infants who received lutein and zeaxanthin supplementation with those who did not (risk ratio (RR) 0.90, 95% confidence interval (CI) 0.66 to 1.24; P = 0.53; I2 = 0%; 4 studies, 532 infants; moderate-certainty evidence). Lutein and zeaxanthin supplementation probably reduces the incidence of ROP stage 3 and above (RR 0.49, 95% CI 0.29 to 0.81; P = 0.005; I2 = 0%; 4 studies, 532 infants; moderate-certainty evidence). No studies assessed the incidence of visual impairment. Lutein and zeaxanthin supplementation may have little or no effect on mortality assessed throughout the NICU stay (RR 0.95, 95% CI 0.42 to 2.17; P = 0.91; I2 = 0%; 4 studies, 470 infants; low-certainty evidence), incidence of IVH (all grades) (RR 0.87, 95% CI 0.44 to 1.75; P = 0.70; I2 = 0%; 4 studies, 483 infants; low-certainty evidence), and incidence of NEC: Bell's stage II or greater (RR 0.87, 95% CI 0.43 to 1.76; P = 0.71; I2 = 0%; 5 studies, 666 infants; low-certainty evidence). No adverse effects were reported in either group.

AUTHORS' CONCLUSIONS

While supplementation with lutein and zeaxanthin from day one of life in preterm infants until discharge probably reduces the incidence of ROP stage 3 and above, it may have little or no effect on the incidence of ROP at any stage, IVH or NEC, or mortality assessed throughout the NICU stay. However, the pooled estimates for these outcomes may change with further rigorously conducted trials. There were no adverse effects reported.

Higher initial doses of surfactant for the prevention or treatment of respiratory distress syndrome in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the effects of higher - compared with standard - initial doses of surfactant in high-risk preterm infants and preterm infants with respiratory distress syndrome.

Acupuncture for procedural pain in newborn infants

RATIONALE

Procedural pain management in newborns, particularly those in neonatal intensive care units (NICUs), presents challenges due to limited safe and effective options. Acupuncture, a Traditional Chinese Medicine practice, has emerged as a potential alternative for pain relief in this population.

OBJECTIVES

To assess the benefits and harms of acupuncture in newborn infants undergoing painful procedures.

SEARCH METHODS

We searched the Cochrane Central Register of Controlled Trials (CENTRAL), PubMed, Embase, and clinical trial registries up to August 2023. We checked the references of included studies and related systematic reviews.

ELIGIBILITY CRITERIA

We included parallel and cross-over randomized controlled trials (RCTs) comparing acupuncture with no treatment or sham treatment; any non-pharmacological treatment; any pharmacological treatment; or one type of acupuncture compared to another type of acupuncture.

OUTCOMES

Our outcomes were: pain scores; harms; parental, family, and caregiver satisfaction with the intervention; use of additional pharmacological intervention for pain relief; episodes of bradycardia/apnea/desaturation; neonatal mortality; mortality during initial hospitalization; intraventricular hemorrhage; late-onset sepsis; duration of hospital stay; major neurodevelopmental disability.

RISK OF BIAS

We used Cochrane's RoB 1 tool for RCTs.

SYNTHESIS METHODS

We conducted meta-analyses using fixed-effect models to calculate risk ratios (RR) and risk differences (RD) with 95% confidence intervals (CI) for dichotomous outcomes, and mean difference (MD) or standardized mean difference (SMD, when combining different pain scales) and standard deviation for continuous outcomes. We summarized the certainty of evidence according to the GRADE approach.

INCLUDED STUDIES

We included 11 RCTs enrolling 852 infants. Five studies compared acupuncture to no treatment or sham treatment; four studies to non-pharmacological treatment (oral sucrose or glucose); and two studies compared acupuncture to other types of Traditional Chinese Medicine treatments, which we refer to as 'type B acupuncture,' such as foot massage or reflexology. No studies compared acupuncture to any pharmacological treatment. We identified four ongoing studies.

SYNTHESIS OF RESULTS

We have listed outcomes reported in at least one study. Acupuncture compared to no treatment or sham treatment Acupuncture may reduce pain assessed during the procedure with any validated scale compared to no intervention (SMD -0.56, 95% CI -0.75 to -0.37; 7 studies, 471 infants; low-certainty evidence). It may result in little to no difference in any harms compared to no intervention (RR 0.35, 95% CI 0.01 to 8.31; 2 studies, 138 infants; low-certainty evidence). Acupuncture compared to any non-pharmacological treatment The evidence is very uncertain about the effect of acupuncture on pain assessed with a validated scale during the procedure compared to non-pharmacological intervention (SMD 0.29, 95% CI 0.04 to 0.54; 4 studies, 267 infants; very low-certainty evidence). Acupuncture may result in little to no difference in any harms compared to non-pharmacological treatment (RR not estimable, RD 0.00, 95% CI -0.03 to 0.03; 3 studies, 247 infants; low-certainty evidence). Acupuncture type A (e.g. penetration of the skin with a needle) compared to acupuncture type B (e.g. foot massage or reflexology) The evidence is very uncertain about the effect of acupressure on pain assessed with any validated scale during the procedure compared to foot massage (SMD 0.05, 95% CI -0.26 to 0.36; 2 studies, 163 infants; very low-certainty evidence).

AUTHORS' CONCLUSIONS

Acupuncture may reduce pain assessed with different scales during the procedure, with little to no difference in any harms, when compared to no intervention. The evidence is very uncertain about the effect of acupuncture on pain assessed with different scales during the procedure when compared to any non-pharmacological treatment; acupuncture may result in little to no difference in any harms. The evidence is very uncertain about the effect of acupressure on pain assessed during the procedure when compared to foot massage or reflexology.

FUNDING

This Cochrane review had no dedicated funding.

REGISTRATION

Protocol (2023): doi.org/10.1002/14651858.CD015894.

Glucagon-like peptide-1 receptor agonists and sodium-glucose co-transporter-2 inhibitors combination therapy for adults with type 2 diabetes mellitus: a network meta-analysis

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: Primary objective To evaluate the relative benefits and harms of therapy with glucagon-like peptide-1 receptor agonists (GLP-1RA), sodium-glucose co-transporter-2 inhibitors (SGLT2i), and their combination in adults with type 2 diabetes mellitus. Secondary objectives To determine the relative rankings of GLP-1RA, SGLT2i, and their combination, according to their comparative efficacy for the critical outcomes identified in this review through a network meta-analysis.

Vaccines for preventing infections in adults with solid tumours

BACKGROUND

Infections are one of the most frequent complications seen in adults with cancer, often arising from the underlying condition or as a result of immunosuppressive treatments. Certain infections (e.g. influenza, pneumococcal disease, and meningococcal disease) may be prevented through vaccination. However, adults with solid tumours may elicit varying immune responses compared to healthy individuals.

OBJECTIVES

To assess the benefits and risks of vaccines for the prevention of infectious diseases in adults with solid tumours.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, two further databases, and two study registries from inception to 2 December 2024 for randomised controlled trials (RCTs) and controlled non-randomised studies of interventions (NRSIs).

SELECTION CRITERIA

We included RCTs evaluating vaccines against the following infectious diseases in adults (≥ 18 years of age) with any diagnosis of solid tumour cancer compared to placebo or no vaccine: pneumococcal disease, Haemophilus influenzae type b disease, meningococcal disease, pertussis, hepatitis B, tetanus, polio, diphtheria, influenza, herpes zoster, and COVID-19. In cases where RCTs were unavailable, we included prospective controlled NRSIs. We excluded live-attenuated vaccines.

DATA COLLECTION AND ANALYSIS

We followed standard Cochrane methodology. Two review authors independently screened search results, extracted data, and assessed the risk of bias (RoB) in the included studies using the Cochrane RoB 2 tool for RCTs and ROBINS-I for NRSIs. We rated the certainty in the evidence using the GRADE approach for the following prioritised outcomes: incidence of infection concerned, all-cause mortality, quality of life, adverse events (AEs) of any grade, serious adverse events (SAEs), localised events at the injection site, and systemic events.

MAIN RESULTS

We included 10 studies (five RCTs and five NRSIs) involving 81,823 adults with solid tumours receiving vaccines to prevent infections with herpes zoster, influenza, or COVID-19. Six studies included participants with varied solid tumours, while two focused on neck and oesophageal cancer or lung cancer. We assessed the RCTs to be at low or moderate risk of bias, whereas most NRSIs were at critical risk of bias due to concerns about confounding. We identified two ongoing studies: one RCT evaluating an influenza vaccine, and one NRSI evaluating COVID-19 vaccines. Twelve studies are awaiting assessment. We did not identify RCTs or NRSIs of vaccines for preventing pneumococcal disease, Haemophilus influenzae type b disease, meningococcal disease, pertussis, hepatitis B, tetanus, polio, or diphtheria compared to placebo or no vaccine. The results from the RCTs are presented below. The results from the NRSIs are detailed in the main text of the review. No study reported quality of life. Vaccines for preventing herpes zoster compared to placebo or no vaccine Three RCTs (3054 participants) evaluated vaccines to prevent herpes zoster. Herpes zoster vaccines decrease the incidence of herpes zoster up to 29.4 months after the final dose (RR 0.37, 95% CI 0.23 to 0.59; 1 RCT, 2678 participants; high-certainty evidence). Herpes zoster vaccines probably make little or no difference to all-cause mortality up to 28 days after the final dose (RR 1.17, 95% CI 0.91 to 1.50; 2 RCTs, 2744 participants; moderate-certainty evidence); make little or no difference to any-grade AEs up to 30 days after final dose (RR 1.02, 95% CI 0.98 to 1.05; 3 RCTs, 2976 participants; high-certainty evidence), and probably make little or no difference in SAEs up to 30 days (RR 1.08, 95% CI 0.93 to 1.24; I² = 0%; 3 RCTs, 2976 participants; moderate-certainty evidence). Vaccines to prevent herpes zoster increase the number of participants with localised events at the injection site compared to placebo or no vaccine (RR 6.81, 95% CI 2.52 to 18.40; 3 RCTs, 2966 participants; high-certainty evidence) and may make little or no difference to the number of participants with systemic events up to 30 days after final dose (RR 1.08, 95% CI 0.77 to 1.50; 3 RCTs, 2966 participants; low-certainty evidence). Vaccines for preventing influenza compared to placebo or no vaccine One RCT (75 participants) evaluated vaccines to prevent influenza. We are uncertain about the effects of influenza vaccines administered prior to surgery on all-cause mortality (RR 1.00, 95% CI 0.07 to 15.33; 1 RCT, 66 participants; very low-certainty evidence), any-grade AEs (RR 1.17, 95% CI 0.89 to 1.54; 1 RCT, 75 participants; very low-certainty evidence), and SAEs (RR 1.46, 95% CI 0.76 to 2.83; 1 RCT, 75 participants; very low-certainty evidence) up to 15 days post-surgery. The RCT did not report the incidence of influenza, localised events at the injection site, or systemic events. Vaccines for preventing COVID-19 compared to placebo or no vaccine One RCT (2256 participants) evaluated vaccines to prevent COVID-19. Participants may have been exposed to the SARS-CoV-2 variants alpha, beta, and gamma. Vaccines to prevent COVID-19 probably decrease the incidence of COVID-19 in participants without previous COVID-19 infection up to six months after the second dose (RR 0.08, 95% CI 0.02 to 0.25; 1 RCT, 2100 participants; moderate-certainty evidence). The COVID-19 vaccines probably increase any-grade AEs (RR 1.99, 95% CI 1.71 to 2.30; 1 RCT, 2328 participants; moderate-certainty evidence). They may have little or no effect on SAEs up to 6 months after the second dose (RR 1.43, 95% CI 0.80 to 2.54; 1 RCT, 2328 participants; low-certainty evidence). The RCT did not report localised events at the injection site or systemic events.

AUTHORS' CONCLUSIONS

In adults with solid tumours, herpes zoster vaccines reduced the incidence of herpes zoster (high-certainty evidence), although localised events at the injection site were more likely to occur (high-certainty evidence). The evidence is very uncertain about the effects of influenza vaccines on all-cause mortality, any-grade AEs, and SAEs (very low-certainty evidence); the incidence of influenza was not measured in the studies. COVID-19 vaccines probably decrease the incidence of COVID-19 in those without prior infection (moderate-certainty evidence) but probably increase any-grade AEs (moderate-certainty evidence). We found no RCTs or NRSIs investigating vaccines for preventing pneumococcal disease, Haemophilus influenzae type b disease, meningococcal disease, pertussis, hepatitis B, tetanus, polio, diphtheria compared to placebo or no vaccine, in adults with solid tumours. Additional research, preferably of RCT design, is necessary to resolve uncertainties.

Training to increase parental participation in the care of infants in neonatal units

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the effects of training to increase parental participation in the care of infants in neonatal units, compared to no training or a different type of training.

Next-generation sequencing (NGS) techniques for pre-symptomatic identification of genetic diseases in newborns

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To evaluate the benefits and harms of using NGS techniques compared to conventional newborn screening alone for pre-symptomatic identification of genetic diseases in newborns.

SECONDARY OBJECTIVES

to explore equity and ethical issues in the application of the new techniques, to inform healthcare decisions by families, carers, and policymakers.

Using artificial intelligence tools to automate data extraction for living evidence syntheses

Living evidence synthesis (LES) involves repeatedly updating a systematic review or meta-analysis at regular intervals to incorporate new evidence into the summary results. It requires a considerable amount of human time investment in the article search, collection, and data extraction phases. Tools exist to automate the retrieval of relevant journal articles, but pulling data out of those articles is currently still a manual process. In this article, we present a proof-of-concept Python program that leverages artificial intelligence (AI) tools (specifically, ChatGPT) to parse a batch of journal articles and extract relevant results, greatly reducing the human time investment in this action without compromising on accuracy. Our program is tested on a set of journal articles that estimate the mean incubation period for COVID-19, an epidemiological parameter of importance for mathematical modelling. We also discuss important limitations related to the total amount of information and rate at which that information can be sent to the AI engine. This work contributes to the ongoing discussion about the use of AI and the role such tools can have in scientific research.

Effect of adrenergic agonist oral decongestants on blood pressure

BACKGROUND

Adrenergic agonist oral decongestants are commonly taken daily over long periods of time to relieve sinus and nasal congestion. The mechanism of action of decongestants potentially increases blood pressure, and these effects may be acute or chronic. However, no systematic reviews to date have comprehensively assessed the chronic blood pressure effects of adrenergic agonist oral decongestants as a drug class, despite their widespread non-prescription availability.

OBJECTIVES

Primary objective To assess the effects of adrenergic agonist oral decongestants on systolic and diastolic blood pressure compared to placebo. Secondary objective To assess the effects of adrenergic agonist oral decongestants on heart rate and withdrawals due to adverse effects.

SEARCH METHODS

The Cochrane Hypertension Information Specialist searched the following databases for randomized controlled trials (RCT) up to July 2024: Cochrane Hypertension Specialised Register, Cochrane Central Register of Controlled Trials (CENTRAL) via Cochrane Register of Studies, Ovid MEDLINE, Ovid Embase, and ClinicalTrials.gov. There were no language, publication year, or publication status restrictions.

SELECTION CRITERIA

We included RCTs of at least seven days' duration with parallel groups (intervention and placebo) comparing adrenergic agonist oral decongestants versus placebo on blood pressure in people aged over six years.

DATA COLLECTION AND ANALYSIS

Two review authors (JJC and MC) independently assessed the trials for inclusion, extracted the data, and assessed the risk of bias from the included trials. In cases where there were disagreements, the third review author (JMW) adjudicated. For any missing or unclear information in the studies, we contacted the study author to request the missing information or seek clarification. The primary outcomes were systolic blood pressure and diastolic blood pressure. Secondary outcomes were heart rate and withdrawal due to adverse effects. We used a fixed-effect model to combine the effect sizes from all studies. We assessed the certainty of the evidence using GRADE.

MAIN RESULTS

Five RCTs randomizing 882 participants met the inclusion criteria. The shortest study duration was one week, and the longest study duration was 24 weeks. These studies measured blood pressure and heart rate after one to seven weeks of taking oral decongestants. The largest study included 568 people, and the smallest study included 18 people. The mean age of participants was 20.0 years, with 326 males and 591 females. The studies were conducted in the USA and Europe; most were set in the USA. Pharmaceutical companies funded three of the five included studies. The findings are relevant to all people using adrenergic agonist oral decongestants for seven days or longer regardless of comorbidities or pre-existing conditions. The adrenergic agonist oral decongestants included were ephedrine (one RCT), pseudoephedrine (one RCT), and phenylpropanolamine (three RCTs). Compared to placebo, adrenergic agonist oral decongestants may have little to no effect on systolic blood pressure, but the evidence is very uncertain (mean difference [MD] 0.91 mmHg, 95% confidence interval [CI] -0.57 to 2.38; 4 RCTs, 784 participants; I2 = 54%; very low-certainty evidence). Adrenergic agonist oral decongestants may have little to no effect on diastolic blood pressure (MD 0.44 mmHg, 95% CI -0.59 to 1.48; 5 RCTs, 882 participants; I2 = 30%; low-certainty evidence). Adrenergic agonist oral decongestants may have little to no effect on heart rate, but the evidence is very uncertain (MD 1.92 beats per minute, 95% CI -0.62 to 4.47; 2 RCTs, 190 participants; I2 = 79%; very low-certainty evidence). Adrenergic agonist oral decongestants may have little to no effect on the number of withdrawals due to adverse effects (risk ratio [RR] 1.20, 95% CI 0.37 to 3.88; 4 RCTs, 806 participants; low-certainty evidence; note 1 study reported withdrawals due to adverse effects but recorded 0 events in both groups. Thus, the RR and CIs were calculated from 3 studies rather than 4). Four studies were at overall high risk of bias, and one study was at overall low risk of bias.

AUTHORS' CONCLUSIONS

In this systematic review, chronic daily intake of adrenergic agonist oral decongestants had little to no effect on blood pressure, heart rate, and withdrawals due to adverse effects. People making decisions about using these medicines should consider the very low certainty of evidence and the theoretical risk of increased blood pressure. Independent trials free from biases that are designed to answer this question are required. The evidence was limited due to the small number of studies, and the blood pressure and heart rate not being measured at the optimal time after drug ingestion.

Publication Type Tagging using Transformer Models and Multi-Label Classification

Indexing articles by their publication type and study design is essential for efficient search and filtering of the biomedical literature, but is understudied compared to indexing by MeSH topical terms. In this study, we leveraged the human-curated publication types and study designs in PubMed to generate a dataset of more than 1.2M articles (titles and abstracts) and used state-of-the-art Transformer-based models for automatic tagging of publication types and study designs. Specifically, we trained PubMedBERT-based models using a multi-label classification approach, and explored undersampling, feature verbalization, and contrastive learning to improve model performance. Our results show that PubMedBERT provides a strong baseline for publication type and study design indexing; undersampling, feature verbalization, and unsupervised constrastive loss have a positive impact on performance, whereas supervised contrastive learning degrades the performance. We obtained the best overall performance with 80% undersampling and feature verbalization (0.632 macro-F1, 0.969 macro-AUC). The model outperformed previous models (MultiTagger) across all metrics and the performance difference was statistically significant (p < 0.001). Despite its stronger performance, the model still has room for improvement and future work could explore features based on full-text as well as model interpretability. We make our data and code available at https://github.com/ScienceNLP-Lab/MultiTagger-v2/tree/main/AMIA.

Harnessing AI for enhanced evidence-based laboratory medicine (EBLM)

The integration of artificial intelligence (AI) into laboratory medicine, is revolutionizing diagnostic accuracy, operational efficiency, and personalized patient care. AI technologies(machine learning, natural language processing and computer vision) advance evidence-based laboratory medicine (EBLM) by automating and optimizing critical processes(formulating clinical questions, conducting literature searches, appraising evidence, and developing clinical guidelines). These reduce the time for systematic reviews, ensuring consistency in appraisal, and enabling real-time updates to guidelines. AI supports personalized medicine by analyzing large datasets, genetic information and electronic health records (EHRs), to tailor diagnostic and treatment plans to patient profiles. Predictive analytics enhance outcomes by leveraging historical data and ongoing monitoring to predict responses and optimize care pathways. Despite the transformative potential, there are challenges. The accuracy, transparency, and explainability of AI algorithms is critical for gaining trust and ensuring ethical deployment. Integration into existing clinical workflows requires collaboration between AI developers and users to ensure seamless user-friendly adoption. Ethical considerations, such as privacy,data security, and algorithmic bias, must also be addressed to mitigate risks and ensure equitable healthcare delivery. Regulatory frameworks, eg. The EU AI Regulation, emphasize transparency, data governance, and human oversight, particularly for high-risk AI systems. The economic and operational benefits are cost savings, improved diagnostic precision, and enhanced patient outcomes. Future trends (federated learning and self-supervised learning), will enhance the scalability and applicability of AI in EBLM, paving the way for a new era of precision medicine. AI in EBLM has the potential to transform healthcare delivery, improve patient outcomes, and advance personalized/precision medicine.

Comparing Digital Versus Face-to-Face Delivery of Systemic Psychotherapy Interventions: Systematic Review and Meta-Analysis of Randomized Controlled Trials

BACKGROUND

As digital mental health delivery becomes increasingly prominent, a solid evidence base regarding its efficacy is needed.

OBJECTIVE

This study aims to synthesize evidence on the comparative efficacy of systemic psychotherapy interventions provided via digital versus face-to-face delivery modalities.

METHODS

We followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for searching PubMed, Embase, Cochrane CENTRAL, CINAHL, PsycINFO, and PSYNDEX and conducting a systematic review and meta-analysis. We included randomized controlled trials comparing mental, behavioral, and somatic outcomes of systemic psychotherapy interventions using self- and therapist-guided digital versus face-to-face delivery modalities. The risk of bias was assessed with the revised Cochrane Risk of Bias tool for randomized trials. Where appropriate, we calculated standardized mean differences and risk ratios. We calculated separate mean differences for nonaggregated analysis.

RESULTS

We screened 3633 references and included 12 articles reporting on 4 trials (N=754). Participants were youths with poor diabetic control, traumatic brain injuries, increased risk behavior likelihood, and parents of youths with anorexia nervosa. A total of 56 outcomes were identified. Two trials provided digital intervention delivery via videoconferencing: one via an interactive graphic interface and one via a web-based program. In total, 23% (14/60) of risk of bias judgments were high risk, 42% (25/60) were some concerns, and 35% (21/60) were low risk. Due to heterogeneity in the data, meta-analysis was deemed inappropriate for 96% (54/56) of outcomes, which were interpreted qualitatively instead. Nonaggregated analyses of mean differences and CIs between delivery modalities yielded mixed results, with superiority of the digital delivery modality for 18% (10/56) of outcomes, superiority of the face-to-face delivery modality for 5% (3/56) of outcomes, equivalence between delivery modalities for 2% (1/56) of outcomes, and neither superiority of one modality nor equivalence between modalities for 75% (42/56) of outcomes. Consequently, for most outcome measures, no indication of superiority or equivalence regarding the relative efficacy of either delivery modality can be made at this stage. We further meta-analytically compared digital versus face-to-face delivery modalities for attrition (risk ratio 1.03, 95% CI 0.52-2.03; P=.93) and number of sessions attended (standardized mean difference -0.11; 95% CI -1.13 to -0.91; P=.83), finding no significant differences between modalities, while CIs falling outside the range of the minimal important difference indicate that equivalence cannot be determined at this stage.

CONCLUSIONS

Evidence on digital and face-to-face modalities for systemic psychotherapy interventions is largely heterogeneous, limiting conclusions regarding the differential efficacy of digital and face-to-face delivery. Nonaggregated and meta-analytic analyses did not indicate the superiority of either delivery condition. More research is needed to conclude if digital and face-to-face delivery modalities are generally equivalent or if-and in which contexts-one modality is superior to another.

TRIAL REGISTRATION

PROSPERO CRD42022335013; https://tinyurl.com/nprder8h.

Cardiovascular training for fatigue in people with cancer

RATIONALE

Cancer-related fatigue (CRF) is the most prevalent and severe symptom among people with cancer. It can be attributed to the cancer itself or to anticancer therapies. CRF affects the individual physically and mentally, and cannot be alleviated by rest. Studies show a positive effect of exercise on CRF.

OBJECTIVES

To evaluate the effects of cardiovascular training on cancer-related fatigue (CRF), quality of life (QoL), adverse events, anxiety, and depression in people with cancer, with regard to their stage of anticancer therapy (before, during, or after), up to 12 weeks, up to six months, or longer, postintervention.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, ClinicalTrials.gov and World Health Organization ICTRP to identify studies that are included in the review. The latest search date was October 2023.

ELIGIBILITY CRITERIA

We included randomised controlled trials (RCTs) evaluating cardiovascular training for CRF or QoL, or both, in people with cancer. Trials were eligible if training was structured, included at least five sessions, and instruction was face-to-face (via video tools or in person). We excluded studies with fewer than 20 randomised participants per group and where only an abstract was available.

OUTCOMES

Our critical outcomes were: short-, medium-, long-term CRF and QoL. Important outcomes were adverse events, and short-, medium-, long-term anxiety and depression.

RISK OF BIAS

We used the Cochrane RoB 1 tool to assess bias in RCTs.

SYNTHESIS METHODS

We used standard Cochrane methodology. We synthesised results for each outcome using meta-analysis where possible (inverse variance or Mantel-Haenszel; random-effects model). We pooled data for the respective assessment periods above. We used GRADE to assess certainty of evidence for each outcome.

INCLUDED STUDIES

We included 23 RCTs with 2135 participants, of whom 96.6% originated from high-income countries; 1101 participants were randomised to cardiovascular training and 1034 to no training. Studies included mostly females who were diagnosed with breast cancer. We also identified 36 ongoing and 12 completed studies that have not yet published (awaiting assessment). We only present findings on CRF, QoL and adverse events. For details regarding anxiety and depression, see full text.

SYNTHESIS OF RESULTS

Cardiovascular training before anticancer therapy versus no training for people with cancer We identified no studies for inclusion in this comparison. Cardiovascular training during anticancer therapy versus no training for people with cancer We included 10 studies (1026 participants); eight studies contributed data to quantitative analyses (860 participants). Cardiovascular training probably reduces short-term CRF slightly (mean difference (MD) 2.85, 95% confidence interval (CI) 1.16 to 4.55, on the Functional Assessment of Cancer Therapy - Fatigue (FACT-F), scale 0 to 52, higher values mean better outcome; minimally important difference (MID) 3; 6 studies, 593 participants) and probably results in little to no difference in short-term QoL (MD 3.56, 95% CI 0.21 to 6.90, on the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire C30 (EORTC QLQ C-30), scale 0 to 100, higher values mean better outcome, MID 10; 6 studies, 612 participants) (both moderate-certainty evidence). We are uncertain about the effects on medium-term CRF (MD 2.67, 95% CI -2.58 to 7.92, on FACT-F; MID 3; 1 study, 62 participants), long-term CRF (MD 0.41, 95% CI -2.24 to 3.05, on FACT-F; MID 3; 2 studies, 230 participants), medium-term QoL (MD 6.79, 95% CI -4.39 to 17.97, on EORTC QLQ C-30; MID 10; 1 study, 62 participants), and long-term QoL (MD 1.51, 95% CI -3.40 to 6.42, on EORTC QLQ C-30; MID 10; 2 studies, 230 participants) (all very low-certainty evidence). For adverse events (any grade and follow-up), we did not perform meta-analysis due to heterogeneous definitions, reporting, and measurement (9 RCTs, 955 participants; very low-certainty evidence). Cardiovascular training after anticancer therapy versus no training for people with cancer We included 13 studies (1109 participants); nine studies contributed data to quantitative analyses (756 participants). We are uncertain about the effects of cardiovascular training on short-term CRF (MD 3.62, 95% CI 0 to 7.13, on FACT-F; MID 3; 6 studies, 497 participants), long-term CRF (MD -0.80, 95% CI -1.72 to 0.13, on the Fatigue Symptom Inventory (FSI), scale 1 to 10, higher values mean worse outcome; MID 1; 2 studies, 262 participants), short-term QoL (MD 3.70, 95% CI -0.14 to 7.41, on the Functional Assessment of Cancer Therapy - General (FACT-G), scale 0 to 108, higher values mean better outcome; MID 4; 8 studies, 642 participants), long-term QoL (MD 3.10, 95% CI -1.12 to 7.32, on FACT-G; MID 4; 1 study, 201 participants), and adverse events (risk ratio (RR) 2.71, 95% CI 0.58 to 12.67; 1 study, 50 participants) (all very low-certainty evidence). There were no data for medium-term CRF and QoL.

AUTHORS' CONCLUSIONS

Moderate-certainty evidence shows that cardiovascular training by people with cancer during their anticancer therapy slightly reduces short-term CRF and results in little to no difference in short-term QoL. We do not know whether cardiovascular training increases or decreases medium-term CRF/QoL, and long-term CRF/QoL. There is very low-certainty evidence (due to heterogeneous definitions, reporting and measurement) evaluating whether the training increases or decreases adverse events. In people with cancer who perform cardiovascular training after anticancer therapy, we are uncertain about the effects on short-term CRF/QoL, long-term CRF/QoL, and adverse events. We identified a lack of evidence concerning cardiovascular training before anticancer therapy and on safety outcomes. The 36 ongoing and 12 completed, but unpublished, studies could help close this gap, and could contribute to improving the effect estimates and certainty.

FUNDING

This Cochrane review was funded by the Federal Ministry of Education and Research of Germany, grant number: FKZ 01KG2017.

REGISTRATION

Protocol available via DOI: 10.1002/14651858.CD015211.

Tumor necrosis factor (TNF) inhibitors for psoriatic arthritis

BACKGROUND

Psoriatic arthritis (PsA) is a chronic arthritis affecting people with psoriasis. If untreated, it may lead to disability. Recommended drugs are non-steroidal anti-inflammatory drugs (NSAIDs), corticosteroids, conventional synthetic disease-modifying anti-rheumatic drugs (csDMARDs), biologic DMARDs (bDMARDs), and targeted synthetic DMARDs (tsDMARDs). Tumour necrosis factor inhibitors (TNFi) are the first choice bDMARDs.

OBJECTIVES

To assess the benefits and harms of TNFi in adults with psoriatic arthritis.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, ClinicalTrials.gov and the World Health Organization trials portal up to 28 March 2024.

SELECTION CRITERIA

We included randomized controlled trials (RCTs) in adults with PsA, comparing TNFi to placebo, physiotherapy, NSAIDs, corticosteroids, and cs/b/tsDMARDs. Major outcomes included clinical improvement, minimal disease activity, physical function, health-related quality of life, radiographic progression, serious adverse events, and withdrawals due to adverse events.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods. The primary comparison was TNFi versus placebo. The primary time point was 12 weeks for clinical improvement; 24 weeks for minimal disease activity, function, quality of life, and radiographic progression; and the end of the trial period for serious adverse events and withdrawals due to adverse events.

MAIN RESULTS

We included 25 RCTs randomizing 7857 participants. Four studies compared TNFi to methotrexate and one to ustekinumab in DMARD-naïve participants. In csDMARD-inadequate responders, 11 studies compared TNFi to placebo; four studies compared TNFi to placebo and ixekizumab, bimekizumab, tofacitinib, or upadacitinib; and three studies compared TNFi to ixekizumab, secukinumab, and ustekinumab. Two studies compared different TNFi. We found no studies with b/tsDMARD-inadequate responders (b/tsDMARD-IR). No studies compared TNFi to NSAIDs, corticosteroids, or physiotherapy. Performance (32%), detection (56%) and reporting (80%) biases were at high or unclear risk across studies. Only one study had a low risk of bias in all domains. We limit reporting to the primary comparison, TNFi versus placebo. DMARD-naïve We found no studies comparing TNFi with placebo in DMARD-naïve participants. csDMARD-inadequate responders TNFi probably result in a large clinical improvement compared to placebo. At 12 weeks, 149/1926 (8%) participants in the placebo group showed a clinical improvement (ACR50) compared to 784/2141 (37%) participants in the TNFi group (risk ratio (RR) 5.63, 95% confidence interval (CI) 3.98 to 7.96; I2 = 65%; 14 studies, 4067 participants; moderate-certainty evidence). TNFi probably result in a higher proportion of participants in minimal disease activity. At 24 weeks, 95/1017 (9%) participants in the placebo group were in minimal disease activity compared to 428/1336 (32%) participants in the TNFi group (RR 3.76, 95% CI 2.39 to 5.92; I2 = 72%; 5 studies, 2353 participants; moderate-certainty evidence). At 24 weeks, TNFi may improve function compared to placebo. The mean change in function from baseline (assessed with the Health Assessment Questionnaire; score from 0 to 3, 0 = no disability; minimal clinically important difference (MCID) = 0.35) was -0.14 points with placebo and 0.33 points lower (0.41 lower to 0.25 lower) with TNFi (I2 = 72%; 8 studies, 2949 participants; low-certainty evidence). TNFi probably result in a clinically important improvement in health-related quality of life. The mean change in quality of life from baseline (assessed with the Short Form 36-item Mental Component Summary questionnaire; score from 0 to 100, 100 = best score; MCID = 1.7) was 2.4 points with placebo and 3.29 points higher (2.18 points higher to 4.40 points higher) with TNFi (I2 = 52%; 8 studies, 2928 participants; moderate-certainty evidence). TNFi probably slightly reduce radiographic progression. The mean change in radiographic progression (assessed with the Sharp/Van der Heijde-PsA score; scale from 0 to 528, 0 = no damage) was 0.25 points with placebo and 0.37 points lower with TNFi (0.48 lower to 0.25 lower) (I2 = 32%; 7 studies, 2478 participants; moderate-certainty evidence) at 24 weeks. We downgraded the evidence to moderate certainty for clinical improvement, minimal disease activity, quality of life, and radiographic progression due to risk of bias. For function, we downgraded the evidence to low certainty for risk of bias and imprecision. TNFi may result in little to no difference in serious adverse events, but may slightly increase withdrawals due to adverse events, compared to placebo. At the end of follow-up: 56/1826 participants (3%) given placebo and 69/1900 (4%) participants given TNFi experienced serious adverse events (RR 1.00, 95% CI 0.70 to 1.42; I2 = 0%; 13 studies, 3866 participants; low-certainty evidence); and 35/1926 (2%) participants given placebo and 65/2140 (3%) given TNFi withdrew due to adverse events (RR 1.53, 95% CI 1.01 to 2.33; I2 = 0%; 14 studies, 4066 participants; low-certainty evidence). We downgraded the evidence to low certainty for risk of bias and imprecision.

AUTHORS' CONCLUSIONS

In csDMARD-inadequate responders, moderate-certainty evidence showed that TNFi probably result in a large clinical improvement, lower disease activity, small decrease in radiographic progression, and better quality of life compared to placebo. Low-certainty evidence showed that TNFi may lead to a slight improvement in physical function compared to placebo. Low-certainty evidence suggested that TNFi may lead to a slight increase in withdrawals due to adverse events, whereas they may result in little to no difference in serious adverse events compared to placebo. No trials assessed TNFi compared to placebo in DMARD-naïve participants or in b/tsDMARD-IR.

Physical rehabilitation approaches for the recovery of function and mobility following stroke

BACKGROUND

Various approaches to physical rehabilitation to improve function and mobility are used after stroke. There is considerable controversy around the relative effectiveness of approaches, and little known about optimal delivery and dose. Some physiotherapists base their treatments on a single approach; others use components from several different approaches.

OBJECTIVES

Primary objective: To determine whether physical rehabilitation is effective for recovery of function and mobility in people with stroke, and to assess if any one physical rehabilitation approach is more effective than any other approach.

SECONDARY OBJECTIVE

To explore factors that may impact the effectiveness of physical rehabilitation approaches, including time after stroke, geographical location of study, intervention dose/duration, intervention provider, and treatment components. Stakeholder involvement: Key aims were to clarify the focus of the review, inform decisions about subgroup analyses, and co-produce statements relating to key implications.

SEARCH METHODS

For this update, we searched the Cochrane Stroke Trials Register (last searched November 2022), CENTRAL (2022, Issue 10), MEDLINE (1966 to November 2022), Embase (1980 to November 2022), AMED (1985 to November 2022), CINAHL (1982 to November 2022), and the Chinese Biomedical Literature Database (to November 2022).

SELECTION CRITERIA

Inclusion criteria: Randomised controlled trials (RCTs) of physical rehabilitation approaches aimed at promoting the recovery of function or mobility in adult participants with a clinical diagnosis of stroke.

EXCLUSION CRITERIA

RCTs of upper limb function or single treatment components.

PRIMARY OUTCOMES

measures of independence in activities of daily living (IADL) and motor function.

SECONDARY OUTCOMES

balance, gait velocity, and length of stay.

DATA COLLECTION AND ANALYSIS

Two independent authors selected studies according to pre-defined eligibility criteria, extracted data, and assessed the risk of bias in the included studies. We used GRADE to assess the certainty of evidence.

MAIN RESULTS

In this review update, we included 267 studies (21,838 participants). Studies were conducted in 36 countries, with half (133/267) in China. Generally, studies were heterogeneous, and often poorly reported. We judged only 14 studies in meta-analyses as at low risk of bias for all domains and, on average, we considered 33% of studies in analyses of primary outcomes at high risk of bias. Is physical rehabilitation more effective than no (or minimal) physical rehabilitation? Compared to no physical rehabilitation, physical rehabilitation may improve IADL (standardised mean difference (SMD) 1.32, 95% confidence interval (CI) 1.08 to 1.56; 52 studies, 5403 participants; low-certainty evidence) and motor function (SMD 1.01, 95% CI 0.80 to 1.22; 50 studies, 5669 participants; low-certainty evidence). There was evidence of long-term benefits for these outcomes. Physical rehabilitation may improve balance (MD 4.54, 95% CI 1.36 to 7.72; 9 studies, 452 participants; low-certainty evidence) and likely improves gait velocity (SMD 0.23, 95% CI 0.05 to 0.42; 18 studies, 1131 participants; moderate-certainty evidence), but with no evidence of long-term benefits. Is physical rehabilitation more effective than attention control? The evidence is very uncertain about the effects of physical rehabilitation, as compared to attention control, on IADL (SMD 0.91, 95% CI 0.06 to 1.75; 2 studies, 106 participants), motor function (SMD 0.13, 95% CI -0.13 to 0.38; 5 studies, 237 participants), and balance (MD 6.61, 95% CI -0.45 to 13.66; 4 studies, 240 participants). Physical rehabilitation likely improves gait speed when compared to attention control (SMD 0.34, 95% CI 0.14 to 0.54; 7 studies, 405 participants; moderate-certainty evidence). Does additional physical rehabilitation improve outcomes? Additional physical rehabilitation may improve IADL (SMD 1.26, 95% CI 0.82 to 1.71; 21 studies, 1972 participants; low-certainty evidence) and motor function (SMD 0.69, 95% CI 0.46 to 0.92; 22 studies, 1965 participants; low-certainty evidence). Very few studies assessed these outcomes at long-term follow-up. Additional physical rehabilitation may improve balance (MD 5.74, 95% CI 3.78 to 7.71; 15 studies, 795 participants; low-certainty evidence) and gait velocity (SMD 0.59, 95% CI 0.26 to 0.91; 19 studies, 1004 participants; low-certainty evidence). Very few studies assessed these outcomes at long-term follow-up. Is any one approach to physical rehabilitation more effective than any other approach? Compared to other approaches, those that focus on functional task training may improve IADL (SMD 0.58, 95% CI 0.29 to 0.87; 22 studies, 1535 participants; low-certainty evidence) and motor function (SMD 0.72, 95% CI 0.21 to 1.22; 20 studies, 1671 participants; very low-certainty evidence) but the evidence in the latter is very uncertain. The benefit was sustained long-term. The evidence is very uncertain about the effect of functional task training on balance (MD 2.16, 95% CI -0.24 to 4.55) and gait velocity (SMD 0.28, 95% CI -0.01 to 0.56). Compared to other approaches, neurophysiological approaches may be less effective than other approaches in improving IADL (SMD -0.34, 95% CI -0.63 to -0.06; 14 studies, 737 participants; low-certainty evidence), and there may be no difference in improving motor function (SMD -0.60, 95% CI -1.32 to 0.12; 13 studies, 663 participants; low-certainty evidence), balance (MD -0.60, 95% CI -5.90 to 6.03; 9 studies, 292 participants; low-certainty evidence), and gait velocity (SMD -0.17, 95% CI -0.62 to 0.27; 16 studies, 630 participants; very low-certainty evidence), but the evidence is very uncertain about the effect on gait velocity. For all comparisons, the evidence is very uncertain about the effects of physical rehabilitation on adverse events and length of hospital stay.

AUTHORS' CONCLUSIONS

Physical rehabilitation, using a mix of different treatment components, likely improves recovery of function and mobility after stroke. Additional physical rehabilitation, delivered as an adjunct to 'usual' rehabilitation, may provide added benefits. Physical rehabilitation approaches that focus on functional task training may be useful. Neurophysiological approaches to physical rehabilitation may be no different from, or less effective than, other physical rehabilitation approaches. Certainty in this evidence is limited due to substantial heterogeneity, with mainly small studies and important differences between study populations and interventions. We feel it is unlikely that any studies published since November 2022 would alter our conclusions. Given the size of this review, future updates warrant consensus discussion amongst stakeholders to ensure the most relevant questions are explored for optimal decision-making.

Home visiting for postpartum depression

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits of home visiting models for postpartum depression amongst mothers of young children, where either the mothers or the children are enrolled in early childhood home visiting programs or interventions. To identify core components essential for a home visiting program to effectively address postpartum depression in mothers of young children. The hypothesized core components of such a program include the priority level given to PPD intervention, the type of home visitors, the intensity of the program, the percentage of BIPOC participants, and the service recipients' baseline level of and risk for depression.

Individualized developmental care interventions for promoting development and preventing morbidity in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of individualized developmental care interventions for promoting development and preventing morbidity in preterm infants.

Air versus supplemental oxygen for resuscitation of term or late preterm infants at birth

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: Primary objective To assess the benefits and harms of air compared with supplemental oxygen for resuscitation of term or late preterm infants at birth in reducing rates of mortality and long-term neurodevelopmental impairment (NDI). Secondary objectives To assess whether the benefits and harms of air compared with supplemental oxygen differ according to different oxygen concentrations, gestational age (GA), whether oxygen was titrated to saturation curves and the income of the study country.

Cycled light in the intensive care unit for preterm and low birth weight infants

BACKGROUND

Preterm and low birth weight infants are at an early stage of development, and do not receive adequate maternal circadian signals. They are often cared for over prolonged periods of hospitalisation in neonatal intensive care units (NICU), where environmental circadian stimuli are lacking. Exposure to artificial light-dark cycles may stimulate the development of the circadian system and improve clinical outcomes. However, it remains uncertain whether cycled light (CL) is preferable to near darkness (ND) or continuous bright light (CBL) in fostering development and maturation, and reducing adverse neonatal health outcomes. This is an update of an earlier Cochrane review, last published in 2016.

OBJECTIVES

To evaluate the benefits and harms of CL in preterm and low birth weight infants compared to ND or CBL.

SEARCH METHODS

We searched CENTRAL, PubMed, Embase, and two trial registries to September 2023. We also checked reference lists, and searched for retractions of included studies.

SELECTION CRITERIA

We included randomised controlled trials (RCTs) or quasi-RCTs in preterm infants (< 37 weeks' postmenstrual age (PMA)), or those with a low birth weight (< 2500 g), admitted and cared for in an NICU or a step-down unit, comparing CL with ND or CBL.

DATA COLLECTION AND ANALYSIS

We used the standard review methods of the Cochrane Neonatal Review Group to assess the methodological quality of studies. We used the fixed-effect model with risk ratio (RR) and mean difference (MD), with their 95% confidence intervals (CIs) for dichotomous data. Our primary outcomes were (1) growth at three and six months' corrected age, (2) major neurodevelopmental disability, and (3) adverse effects. Our secondary outcomes were (4) retinopathy of prematurity, (5) duration of initial hospitalisation, (6) duration of oxygen treatment, and (7) parent satisfaction. We used GRADE to assess the certainty of evidence for each outcome.

MAIN RESULTS

We included 20 studies with 1633 infants. Data for meta-analysis were available for 11 studies (1126 infants). One study with multiple arms was included in both comparisons. We rated the overall risk of bias at the study level as high or unclear for all 20 studies that had one or several unclear or high risk of bias judgements across the domains. Cycled light versus dimmed light or near darkness (10 studies) The evidence is very uncertain about the effect of cycled light compared to dimmed light (reduction of illumination levels) or near darkness on weight at three months (MD 24.79, 95% CI -262.33 to 311.91; 2 studies, 187 infants; very low-certainty evidence), and weight at six months (MD 202, 95% CI -109.68 to 513.68; 1 study, 147 infants; very low-certainty evidence). The studies did not report any data for major neurodevelopmental disability. No data are available for adverse effects; it is uncertain if the absence of adverse effects is because none occurred, or because they were not identified and recorded. The evidence is very uncertain about the effect of cycled light compared to dimmed light or near darkness on the likelihood of developing retinopathy of prematurity of any stage (RR 0.89, 95% CI 0.76 to 1.03; 3 studies, 307 infants; very low-certainty evidence), and severe retinopathy of prematurity of stage 3 or higher (RR 0.98, 95% CI 0.59 to 1.61; 4 studies, 454 infants; very low-certainty evidence). Cycled light compared to dimmed light or near darkness may have little to no effect on the duration of initial hospitalisation (MD -3.04, 95% CI -7.86 to 1.78; 5 studies, 550 infants; very low-certainty evidence), but the evidence is very uncertain. Cycled light versus continuous bright light (11 studies) No data are available on the following primary outcomes, as no studies reported them: growth at three and six months' corrected age, major neurodevelopmental disability, and adverse effects. It is uncertain if the absence of adverse effects is because none occurred or because they were not identified and recorded. No data are available on retinopathy of prematurity, as no studies reported it. Cycled light compared to continuous bright light may reduce the duration of initial hospitalisation, but the evidence is very uncertain (MD -9.86, 95% CI -10.09 to -9.63; 5 studies, 499 infants; very low-certainty evidence).

AUTHORS' CONCLUSIONS

Despite identifying 20 studies, we remain uncertain about the effect of CL compared to ND or CBL on all outcomes of interest in this review. In addition, a few critical outcomes were not reported by any of the included studies. The evidence remains uncertain about whether CL is the right choice in the NICU. The physician should always weigh the benefits and risks, based on the effects of the different options in the specific setting.

Clustering of care activities for promoting development and preventing morbidity in hospitalized preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of clustering of care activities for promoting development and preventing morbidity in preterm infants.

Financial incentives for family members of hospitalized neonates for improving family presence

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To evaluate the benefits and harms of financial incentives for improving family engagement on family members of neonates receiving hospitalized care.

Strategies for enhancing the implementation of school-based policies or practices targeting diet, physical activity, obesity, tobacco or alcohol use

BACKGROUND

A range of school-based interventions are effective in improving student diet and physical activity (e.g. school food policy interventions and classroom physical activity interventions), and reducing obesity, tobacco use and/or alcohol use (e.g. tobacco control programmes and alcohol education programmes). However, schools are frequently unsuccessful in implementing such evidence-based interventions.

OBJECTIVES

The primary review objective is to evaluate the effectiveness of strategies aiming to improve school implementation of interventions to address students' (aged 5 to 18 years) diet, physical activity, obesity, tobacco use and/or alcohol use. The secondary objectives are to: 1. determine whether the effects are different based on the characteristics of the intervention including school type and the health behaviour or risk factor targeted by the intervention; 2. describe any unintended consequences and adverse effects of strategies on schools, school staff or students; and 3. describe the cost or cost-effectiveness of strategies.

SEARCH METHODS

We searched CENTRAL, MEDLINE (Ovid), Embase (Ovid), five additional databases, the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), and the US National Institutes of Health registry (clinicaltrials.gov). The latest search was between 1 May 2021 and 30 June 2023 to identify any relevant trials published since the last published review.

SELECTION CRITERIA

We defined 'implementation' as the use of strategies to adopt and integrate evidence-based health interventions and to change practice patterns within specific settings. We included any randomised controlled trial (RCT) or cluster-RCT conducted on any scale, in a school setting, with a parallel control group that compared a strategy to improve the implementation of policies or practices to address diet, physical activity, obesity, tobacco use and/or alcohol use by students (aged 5 to 18 years) to no active implementation strategy (i.e. no intervention, inclusive of usual practice, minimal support) or a different implementation strategy.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods. Given the large number of outcomes reported, we selected and included the effects of a single outcome measure for each trial for the primary outcome using a decision hierarchy (i.e. continuous over dichotomous, most valid, total score over subscore). Where possible, we calculated standardised mean differences (SMDs) to account for variable outcome measures with 95% confidence intervals (CI). We conducted meta-analyses using a random-effects model. Where we could not combine data in meta-analysis, we followed recommended Cochrane methods and reported results in accordance with 'Synthesis without meta-analysis' (SWiM) guidelines. We conducted assessments of risk of bias and evaluated the certainty of evidence (GRADE approach) using Cochrane procedures.

MAIN RESULTS

We included an additional 14 trials in this update, bringing the total number of included trials in the review to 39 trials with 83 trial arms and 6489 participants. Of these, the majority were conducted in Australia and the USA (n = 15 each). Nine were RCTs and 30 were cluster-RCTs. Twelve trials tested strategies to implement healthy eating practices; 17 physical activity, two tobacco, one alcohol, and seven a combination of risk factors. All trials used multiple implementation strategies, the most common being educational materials, educational meetings, and education outreach visits, or academic detailing. Of the 39 included trials, we judged 26 as having high risks of bias, 11 as having some concerns, and two as having low risk of bias across all domains. Pooled analyses found, relative to a control (no active implementation strategy), the use of implementation strategies probably results in a large increase in the implementation of interventions in schools (SMD 0.95, 95% CI 0.71, 1.19; I2 = 78%; 30 trials, 4912 participants; moderate-certainty evidence). This is equivalent to a 0.76 increase in the implementation of seven physical activity intervention components when the SMD is re-expressed using an implementation measure from a selected included trial. Subgroup analyses by school type and targeted health behaviour or risk factor did not identify any differential effects, and only one study was included that was implemented at scale. Compared to a control (no active implementation strategy), no unintended consequences or adverse effects of interventions were identified in the 11 trials that reported assessing them (1595 participants; moderate-certainty evidence). Nine trials compared costs between groups with and without an implementation strategy and the results of these comparisons were mixed (2136 participants; low-certainty evidence). A lack of consistent terminology describing implementation strategies was an important limitation of the review.

AUTHORS' CONCLUSIONS

We found the use of implementation strategies probably results in large increases in implementation of interventions targeting healthy eating, physical activity, tobacco and/or alcohol use. While the effectiveness of individual implementation strategies could not be determined, such examination will likely be possible in future updates as data from new trials can be synthesised. Such research will further guide efforts to facilitate the translation of evidence into practice in this setting. The review will be maintained as a living systematic review.

Infant positioning for promoting development and preventing morbidity in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of infant positioning, nesting, and swaddling for promoting development and preventing morbidity in preterm infants.

Metastatic Versus Localized Disease as Inclusion Criteria That Can Be Automatically Extracted From Randomized Controlled Trials Using Natural Language Processing

PURPOSE

Extracting inclusion and exclusion criteria in a structured, automated fashion remains a challenge to developing better search functionalities or automating systematic reviews of randomized controlled trials in oncology. The question "Did this trial enroll patients with localized disease, metastatic disease, or both?" could be used to narrow down the number of potentially relevant trials when conducting a search.

METHODS

Six hundred trials from high-impact medical journals were classified depending on whether they allowed for the inclusion of patients with localized and/or metastatic disease. Five hundred trials were used to develop and validate three different models, with 100 trials being stored away for testing. The test set was also used to evaluate the performance of GPT-4o in the same task.

RESULTS

In the test set, a rule-based system using regular expressions achieved F1 scores of 0.72 for the prediction of whether the trial allowed for the inclusion of patients with localized disease and 0.77 for metastatic disease. A transformer-based machine learning (ML) model achieved F1 scores of 0.97 and 0.88, respectively. A combined approach where the rule-based system was allowed to over-rule the ML model achieved F1 scores of 0.97 and 0.89, respectively. GPT-4o achieved F1 scores of 0.87 and 0.92, respectively.

CONCLUSION

Automatic classification of cancer trials with regard to the inclusion of patients with localized and/or metastatic disease is feasible. Turning the extraction of trial criteria into classification problems could, in selected cases, improve text-mining approaches in evidence-based medicine. Increasingly large language models can reduce or eliminate the need for previous training on the task at the expense of increased computational power and, in turn, cost.

Resistance training for fatigue in people with cancer

BACKGROUND

Cancer-related fatigue (CRF) is one of the most common symptoms associated with cancer and its treatment. Different types of exercise have demonstrated beneficial effects on CRF. Previous evidence syntheses provided promising but inconclusive results when focusing on the effects of resistance training.

OBJECTIVES

To evaluate the effects of resistance training on CRF in people with cancer and, specifically, to compare the effects of resistance training with no training on CRF at: different periods of treatment in relation to anticancer therapy (before, during, or after anticancer therapy); different periods of assessment (up to 12 weeks after the intervention, between more than 12 weeks and less than six months after the intervention, or six months or longer after the intervention). Moreover, we wanted to compare the effects of resistance training with no training on quality of life (QoL), adverse events, depression, and anxiety.

SEARCH METHODS

We performed an extensive literature search in eight databases including CENTRAL, Medline, and Embase in October 2023. We searched trial registries for ongoing studies, and we integrated results from update searches of previously published Cochrane reviews.

SELECTION CRITERIA

We included randomised controlled trials (RCTs) that compared resistance training with no training in adults with any type of cancer who received resistance training initiated before, during, or after anticancer therapy. Eligible RCTs needed to evaluate CRF or QoL. Resistance training had to be structured, last for at least five sessions, and include face-to-face instruction. We excluded studies that randomised fewer than 20 participants per group.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methodology. For analyses, we pooled short-term, medium-term, and long-term effects (i.e. up to 12 weeks, between more than 12 weeks and less than six months, and six months or longer, after the intervention). We assessed risk of bias and certainty of the evidence using Cochrane's risk of bias tool (RoB 1), and the GRADE approach, respectively.

MAIN RESULTS

We included 21 RCTs with a total of 2221 participants, with diverse types of cancer, who received resistance training initiated during (14 studies), or after (7 studies) anticancer therapy. None of the studies investigated the effects of resistance training initiated before anticancer therapy. Here, we present the results on CRF, QoL, and adverse events. Results on depression and anxiety are reported in the full review. Resistance training during anticancer therapy Resistance training probably has a beneficial effect compared with no training on short-term CRF (mean difference (MD) on Functional Assessment of Chronic Illness Therapy - Fatigue scale (FACIT-Fatigue) 3.90, 95% confidence interval (CI) 1.30 to 6.51; scale from 0 to 52, higher values mean better outcome, minimal important difference (MID) 3; 12 RCTs, 1120 participants; moderate-certainty evidence). The evidence is very uncertain about the effect of resistance training compared with no training on medium-term CRF (MD on Multidimensional Fatigue Inventory -8.33, 95% CI -18.34 to 1.68; scale from 20 to 100, higher values mean worse outcome, MID 11.5; 1 RCT, 47 participants; very low-certainty evidence). The evidence is very uncertain about the effect of resistance training compared with no training on long-term CRF (MD on FACIT-Fatigue -0.70, 95% CI -4.16 to 2.76; 1 RCT, 133 participants; very low-certainty evidence). Resistance training may have a small beneficial effect compared with no training on short-term QoL (MD on EORTC QoL Questionnaire C30 - global health (QLQ-C30) 4.93, 95% CI 2.01 to 7.85; scale from 0 to 100, higher values mean better outcome, MID 10; 12 RCTs, 1117 participants; low-certainty evidence). The evidence is very uncertain about the effect of resistance training compared with no training on medium-term QoL (MD on QLQ-C30 6.48, 95% CI -4.64 to 17.60; 1 RCT, 42 participants; very low-certainty evidence). The evidence is very uncertain about the effect of resistance training compared with no training on long-term QoL (MD on Functional Assessment of Cancer Therapy - Anemia (FACT-An) 0.50, 95% CI -8.46 to 9.46; scale from 0 to 188; higher values mean better outcome, MID 7; 1 RCT, 133 participants; very low-certainty evidence). Only two RCTs (116 participants) reported data on adverse events for both the resistance training and the control arm. The evidence is very uncertain about the effect of resistance training compared with no training on the occurrence of adverse events (very low-certainty evidence). Resistance training after anticancer therapy The evidence is very uncertain about the effect of resistance training compared with no training on short-term CRF (MD on Chalder Fatigue Scale -0.27, 95% CI -2.11 to 1.57; scale from 0 to 33, higher values mean worse outcome, MID 2.3; 3 RCTs, 174 participants; very low-certainty evidence). Resistance training may have a small beneficial effect or no effect compared with no training on short-term QoL (MD on QLQ-C30 3.87, 95% CI -1.22 to 8.97; 4 RCTs, 243 participants; low-certainty evidence). None of the studies reported data on medium-, or long-term effects on CRF or QoL. Only three RCTs (238 participants) reported data on adverse events for both the resistance training and the control arm. The evidence is very uncertain about the effect of resistance training compared with no training on the occurrence of adverse events (very low-certainty evidence).

AUTHORS' CONCLUSIONS

Our review demonstrates beneficial effects of resistance training during anticancer therapy compared with no training on short-term CRF and QoL for people with cancer. Resistance training after anticancer therapy may also have a small beneficial effect on short-term QoL. Data on medium-, and long-term effects are sparse. In order to facilitate evidence syntheses beyond a narrative report of the data, investigators of resistance training programmes should report adverse events more consistently and completely for all study arms, including control groups.

Methylxanthines for the prevention or treatment of intermittent hypoxemia or respiratory insufficiency in late preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of methylxanthines in preventing or treating intermittent hypoxemia or respiratory insufficiency in late preterm infants.

Minimizing blood sampling in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of different strategies to minimize blood sampling in preterm infants.

Semi-automated title-abstract screening using natural language processing and machine learning

BACKGROUND

Title-abstract screening in the preparation of a systematic review is a time-consuming task. Modern techniques of natural language processing and machine learning might allow partly automatization of title-abstract screening. In particular, clear guidance on how to proceed with these techniques in practice is of high relevance.

METHODS

This paper presents an entire pipeline how to use natural language processing techniques to make the titles and abstracts usable for machine learning and how to apply machine learning algorithms to adequately predict whether or not a publication should be forwarded to full text screening. Guidance for the practical use of the methodology is given.

RESULTS

The appealing performance of the approach is demonstrated by means of two real-world systematic reviews with meta analysis.

CONCLUSIONS

Natural language processing and machine learning can help to semi-automatize title-abstract screening. Different project-specific considerations have to be made for applying them in practice.

Nasal continuous positive airway pressure immediately after extubation for preventing morbidity in preterm infants

BACKGROUND

Preterm infants who are extubated following a period of invasive ventilation via an endotracheal tube are at risk of developing respiratory failure, leading to reintubation. This may be due to apnoea, respiratory acidosis, or hypoxia. Historically, preterm infants were extubated to head box oxygen or low-flow nasal cannulae. Support with non-invasive pressure might help improve rates of successful extubation in preterm infants by stabilising the upper airway, improving lung function, and reducing apnoea. This is an update of a review first published in 1997 and last updated in 2003.

OBJECTIVES

To determine whether nasal continuous positive airway pressure (NCPAP), applied immediately after extubation of preterm infants, reduces the incidence of extubation failure and the need for additional ventilatory support, without clinically important adverse events.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, and trial registries on 22 September 2023 using a revised strategy. We searched conference abstracts and the reference lists of included studies and relevant systematic reviews.

SELECTION CRITERIA

Eligible trials employed random or quasi-random allocation of preterm infants undergoing extubation. Eligible comparisons were NCPAP (delivered by any device and interface) versus head box oxygen, extubation to room air, or any other form of low-pressure supplemental oxygen. We grouped the comparators under the term no continuous positive airway pressure (no CPAP).

DATA COLLECTION AND ANALYSIS

Two review authors independently assessed the risk of bias and extracted data from the included studies. Where studies were sufficiently similar, we performed a meta-analysis, calculating risk ratios (RRs) with their 95% confidence intervals (CIs) for dichotomous data. For the primary outcomes that showed an effect, we calculated the number needed to treat for an additional beneficial outcome (NNTB). We used the GRADE approach to assess the certainty of the evidence for clinically important outcomes.

MAIN RESULTS

We included nine trials (with 726 infants) in the quantitative synthesis of this updated review. Eight studies were conducted in high-income countries between 1982 and 2005. One study was conducted in Chile, which was classified as upper-middle income at the time of the study. All studies used head box oxygen in the control arm. Risk of bias was generally low. However, due to the inherent nature of the intervention, no studies incorporated blinding. Consequently, the neonatal intensive care unit staff were aware of the assigned group for each infant, and we judged all studies at high risk of performance bias. However, we assessed blinding of the outcome assessor (detection bias) as low risk for seven studies because they used objective criteria to define both primary outcomes. NCPAP compared with no CPAP may reduce the risk of extubation failure (RR 0.62, 95% CI 0.51 to 0.76; risk difference (RD) -0.17, 95% -0.23 to -0.10; NNTB 6, 95% CI 4 to 10; I2 = 55%; 9 studies, 726 infants; low-certainty evidence) and endotracheal reintubation (RR 0.79, 95% 0.64 to 0.98; RD -0.07, 95% CI -0.14 to -0.01; NNTB 15, 95% CI 8 to 100; I2 = 65%; 9 studies; 726 infants; very low-certainty evidence), though the evidence for endotracheal reintubation is very uncertain. NCPAP compared with no CPAP may have little or no effect on bronchopulmonary dysplasia, but the evidence is very uncertain (RR 0.89, 95% CI 0.47 to 1.68; RD -0.03, 95% CI -0.22 to 0.15; 1 study, 92 infants; very low-certainty evidence). No study reported neurodevelopmental outcomes.

AUTHORS' CONCLUSIONS

NCPAP may be more effective than no CPAP in preventing extubation failure in preterm infants if applied immediately after extubation from invasive mechanical ventilation. We are uncertain whether it can reduce the risk of reintubation or bronchopulmonary dysplasia. We have no information on long-term neurodevelopmental outcomes. Although there is only low-certainty evidence for the effectiveness of NCPAP immediately after extubation in preterm infants, we consider there is no need for further research on this intervention, which has become standard practice.

Vestibular stimulation for promoting development and preventing morbidity in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of vestibular stimulation compared to standard care or non-vestibular stimulation for physical and neurological development in preterm infants. To assess whether the effects of vestibular stimulation differ according to gestational age at birth; the type, frequency, and duration of the intervention; and settings, such as the country where the study is conducted.

Shorter versus longer duration antibiotic regimens for treatment of suspected neonatal sepsis

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the safety and effectiveness of shorter versus longer duration antibiotic regimens for the treatment of suspected neonatal sepsis.

Nutritional therapy for reducing disability and improving activities of daily living in people after stroke

BACKGROUND

Stroke patients often face disabilities that significantly interfere with their daily lives. Poor nutritional status is a common issue amongst these patients, and malnutrition can severely impact their functional recovery post-stroke. Therefore, nutritional therapy is crucial in managing stroke outcomes. However, its effects on disability, activities of daily living (ADL), and other critical outcomes have not been fully explored.

OBJECTIVES

To evaluate the effects of nutritional therapy on reducing disability and improving ADL in patients after stroke.

SEARCH METHODS

We searched the trial registers of the Cochrane Stroke Group, CENTRAL, MEDLINE (from 1946), Embase (from 1974), CINAHL (from 1982), and AMED (from 1985) to 19 February 2024. We also searched trials and research registries (ClinicalTrials.gov, World Health Organization International Clinical Trials Registry Platform) and reference lists of articles.

SELECTION CRITERIA

We included randomised controlled trials (RCTs) that compared nutritional therapy with placebo, usual care, or one type of nutritional therapy in people after stroke. Nutritional therapy was defined as the administration of supplemental nutrients, including energy, protein, amino acids, fatty acids, vitamins, and minerals, through oral, enteral, or parenteral methods. As a comparator, one type of nutritional therapy refers to all forms of nutritional therapies, excluding the specific nutritional therapy defined for use in the intervention group.

DATA COLLECTION AND ANALYSIS

We used Cochrane's Screen4Me workflow to assess the initial search results. Two review authors independently screened references that met the inclusion criteria, extracted data, and assessed the risk of bias and the certainty of the evidence using the GRADE approach. We calculated the mean difference (MD) or standardised mean difference (SMD) for continuous data and the odds ratio (OR) for dichotomous data, with 95% confidence intervals (CIs). We assessed heterogeneity using the I2 statistic. The primary outcomes were disability and ADL. We also assessed gait, nutritional status, all-cause mortality, quality of life, hand and leg muscle strength, cognitive function, physical performance, stroke recurrence, swallowing function, neurological impairment, and the development of complications (adverse events) as secondary outcomes.

MAIN RESULTS

We identified 52 eligible RCTs involving 11,926 participants. Thirty-six studies were conducted in the acute phase, 10 in the subacute phase, three in the acute and subacute phases, and three in the chronic phase. Twenty-three studies included patients with ischaemic stroke, three included patients with haemorrhagic stroke, three included patients with subarachnoid haemorrhage (SAH), and 23 included patients with ischaemic or haemorrhagic stroke including SAH. There were 25 types of nutritional supplements used as an intervention. The number of studies that assessed disability and ADL as outcomes were nine and 17, respectively. For the intervention using oral energy and protein supplements, which was a primary intervention in this review, six studies were included. The results for the seven outcomes focused on (disability, ADL, body weight change, all-cause mortality, gait speed, quality of life, and incidence of complications (adverse events)) were as follows: There was no evidence of a difference in reducing disability when 'good status' was defined as an mRS score of 0 to 2 (for 'good status': OR 0.97, 95% CI 0.86 to 1.10; 1 RCT, 4023 participants; low-certainty evidence). Oral energy and protein supplements may improve ADL as indicated by an increase in the FIM motor score, but the evidence is very uncertain (MD 8.74, 95% CI 5.93 to 11.54; 2 RCTs, 165 participants; very low-certainty evidence). Oral energy and protein supplements may increase body weight, but the evidence is very uncertain (MD 0.90, 95% CI 0.23 to 1.58; 3 RCTs, 205 participants; very low-certainty evidence). There was no evidence of a difference in reducing all-cause mortality (OR 0.57, 95% CI 0.14 to 2.28; 2 RCTs, 4065 participants; low-certainty evidence). For gait speed and quality of life, no study was identified. With regard to incidence of complications (adverse events), there was no evidence of a difference in the incidence of infections, including pneumonia, urinary tract infections, and septicaemia (OR 0.68, 95% CI 0.20 to 2.30; 1 RCT, 42 participants; very low-certainty evidence). The intervention was associated with an increased incidence of diarrhoea compared to usual care (OR 4.29, 95% CI 1.98 to 9.28; 1 RCT, 4023 participants; low-certainty evidence) and the occurrence of hyperglycaemia or hypoglycaemia (OR 15.6, 95% CI 4.84 to 50.23; 1 RCT, 4023 participants; low-certainty evidence).

AUTHORS' CONCLUSIONS

We are uncertain about the effect of nutritional therapy, including oral energy and protein supplements and other supplements identified in this review, on reducing disability and improving ADL in people after stroke. Various nutritional interventions were assessed for the outcomes in the included studies, and almost all studies had small sample sizes. This led to challenges in conducting meta-analyses and reduced the precision of the evidence. Moreover, most of the studies had issues with the risk of bias, especially in terms of the absence of blinding and unclear information. Regarding adverse events, the intervention with oral energy and protein supplements was associated with a higher number of adverse events, such as diarrhoea, hyperglycaemia, and hypoglycaemia, compared to usual care. However, the quality of the evidence was low. Given the low certainty of most of the evidence in our review, further research is needed. Future research should focus on targeted nutritional interventions to reduce disability and improve ADL based on a theoretical rationale in people after stroke and there is a need for improved methodology and reporting.

Data-driven rapid detection of Helicobacter pylori infection through machine learning with limited laboratory parameters in Chinese primary clinics

Background

Helicobacter pylori (H. pylori) is a significant global health concern, posing a high risk for gastric cancer. Conventional diagnostic and screening approaches are inaccessible, invasive, inaccurate, time-consuming, and expensive in primary clinics.

Objective

This study aims to apply machine learning (ML) models to detect H. pylori infection using limited laboratory parameters from routine blood tests and to investigate the association of these biomarkers with clinical outcomes in primary clinics.

Methods

A retrospective analysis with three ML and five ensemble models was conducted on 1409 adults from Hubei Provincial Hospital of Traditional Chinese Medicine. evaluating twenty-three blood test parameters and using theC 14 urea breath test as the gold standard for diagnosing H. pylori infection.

Results

In our comparative study employing three different feature selection strategies, Random Forest (RF) model exhibited superior performance over other ML and ensemble models. Multiple evaluation metrics underscored the optimal performance of the RF model (ROC = 0.951, sensitivity = 0.882, specificity = 0.906, F1 = 0.906, accuracy = 0.894, PPV = 0.908, NPV = 0.880) without feature selection. Key biomarkers identified through importance ranking and shapley additive Explanations (SHAP) analysis using the RF model without feature selection include White Blood Cell Count (WBC), Mean Platelet Volume (MPV), Hemoglobin (Hb), Red Blood Cell Count (RBC), Platelet Crit (PCT), and Platelet Count (PLC). These biomarkers were found to be significantly associated with the presence of H. pylori infection, reflecting the immune response and inflammation levels.

Conclusion

Abnormalities in key biomarkers could prompt clinical workers to consider H. pylori infection. The RF model effectively identifies H. pylori infection using routine blood tests, offering potential for clinical application in primary clinics. This ML approach can enhance diagnosis and screening, reducing medical burdens and reliance on invasive diagnostics.

Olfactory stimulation for promoting development and preventing morbidity in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To evaluate the benefits and harms of olfactory stimulation with different odorants in the NICU for promoting development and preventing morbidity in preterm infants.

Educational and psychological interventions for managing atopic dermatitis (eczema)

BACKGROUND

Atopic dermatitis (eczema), can have a significant impact on well-being and quality of life for affected people and their families. Standard treatment is avoidance of triggers or irritants and regular application of emollients and topical steroids or calcineurin inhibitors. Thorough physical and psychological assessment is central to good-quality treatment. Overcoming barriers to provision of holistic treatment in dermatological practice is dependent on evaluation of the efficacy and economics of both psychological and educational interventions in this participant group. This review is based on a previous Cochrane review published in 2014, and now includes adults as well as children.

OBJECTIVES

To assess the clinical outcomes of educational and psychological interventions in children and adults with atopic dermatitis (eczema) and to summarise the availability and principal findings of relevant economic evaluations.

SEARCH METHODS

We searched the Cochrane Skin Specialised Register, CENTRAL, MEDLINE, Embase, APA PsycINFO and two trials registers up to March 2023. We checked the reference lists of included studies and related systematic reviews for further references to relevant randomised controlled trials (RCTs) and contacted experts in the field to identify additional studies. We searched NHS Economic Evaluation Database, MEDLINE and Embase for economic evaluations on 8 June 2022.

SELECTION CRITERIA

Randomised, cluster-randomised and cross-over RCTs that assess educational and psychological interventions for treating eczema in children and adults.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods, with GRADE to assess the certainty of the evidence for each outcome. Primary outcomes were reduction in disease severity, as measured by clinical signs, patient-reported symptoms and improvement in health-related quality-of-life (HRQoL) measures. Secondary outcomes were improvement in long-term control of symptoms, improvement in psychological well-being, improvement in standard treatment concordance and adverse events. We assessed short- (up to 16 weeks after treatment) and long-term time points (more than 16 weeks).

MAIN RESULTS

We included 37 trials (6170 participants). Most trials were conducted in high-income countries (34/37), in outpatient settings (25/37). We judged three trials to be low risk of bias across all domains. Fifteen trials had a high risk of bias in at least one domain, mostly due to bias in measurement of the outcome. Trials assessed interventions compared to standard care. Individual educational interventions may reduce short-term clinical signs (measured by SCORing Atopic Dermatitis (SCORAD); mean difference (MD) -5.70, 95% confidence interval (CI) -9.39 to -2.01; 1 trial, 30 participants; low-certainty evidence) but patient-reported symptoms, HRQoL, long-term eczema control and psychological well-being were not reported. Group education interventions probably reduce clinical signs (SCORAD) both in the short term (MD -9.66, 95% CI -19.04 to -0.29; 3 studies, 731 participants; moderate-certainty evidence) and the long term (MD -7.22, 95% CI -11.01 to -3.43; 3 studies, 1424 participants; moderate-certainty evidence) and probably reduce long-term patient-reported symptoms (SMD -0.47 95% CI -0.60 to -0.33; 2 studies, 908 participants; moderate-certainty evidence). They may slightly improve short-term HRQoL (SMD -0.19, 95% CI -0.36 to -0.01; 4 studies, 746 participants; low-certainty evidence), but may make little or no difference to short-term psychological well-being (Perceived Stress Scale (PSS); MD -2.47, 95% CI -5.16 to 0.22; 1 study, 80 participants; low-certainty evidence). Long-term eczema control was not reported. We don't know whether technology-mediated educational interventions could improve short-term clinical signs (SCORAD; 1 study; 29 participants; very low-certainty evidence). They may have little or no effect on short-term patient-reported symptoms (Patient Oriented Eczema Measure (POEM); MD -0.76, 95% CI -1.84 to 0.33; 2 studies; 195 participants; low-certainty evidence) and probably have little or no effect on short-term HRQoL (MD 0, 95% CI -0.03 to 0.03; 2 studies, 430 participants; moderate-certainty evidence). Technology-mediated education interventions probably slightly improve long-term eczema control (Recap of atopic eczema (RECAP); MD -1.5, 95% CI -3.13 to 0.13; 1 study, 232 participants; moderate-certainty evidence), and may improve short-term psychological well-being (MD -1.78, 95% CI -2.13 to -1.43; 1 study, 24 participants; low-certainty evidence). Habit reversal treatment may reduce short-term clinical signs (SCORAD; MD -6.57, 95% CI -13.04 to -0.1; 1 study, 33 participants; low-certainty evidence) but we are uncertain about any effects on short-term HRQoL (Children's Dermatology Life Quality Index (CDLQI); 1 study, 30 participants; very low-certainty evidence). Patient-reported symptoms, long-term eczema control and psychological well-being were not reported. We are uncertain whether arousal reduction therapy interventions could improve short-term clinical signs (Eczema Area and Severity Index (EASI); 1 study, 24 participants; very low-certainty evidence) or patient-reported symptoms (visual analogue scale (VAS); 1 study, 18 participants; very low-certainty evidence). Arousal reduction therapy may improve short-term HRQoL (Dermatitis Family Impact (DFI); MD -2.1, 95% CI -4.41 to 0.21; 1 study, 91 participants; low-certainty evidence) and psychological well-being (PSS; MD -1.2, 95% CI -3.38 to 0.98; 1 study, 91 participants; low-certainty evidence). Long-term eczema control was not reported. No studies reported standard care compared with self-help psychological interventions, psychological therapies or printed education; or adverse events. We identified two health economic studies. One found that a 12-week, technology-mediated, educational-support programme may be cost neutral. The other found that a nurse practitioner group-education intervention may have lower costs than standard care provided by a dermatologist, with comparable effectiveness.

AUTHORS' CONCLUSIONS

In-person, individual education, as an adjunct to conventional topical therapy, may reduce short-term eczema signs compared to standard care, but there is no information on eczema symptoms, quality of life or long-term outcomes. Group education probably reduces eczema signs and symptoms in the long term and may also improve quality of life in the short term. Favourable effects were also reported for technology-mediated education, habit reversal treatment and arousal reduction therapy. All favourable effects are of uncertain clinical significance, since they may not exceed the minimal clinically important difference (MCID) for the outcome measures used (MCID 8.7 points for SCORAD, 3.4 points for POEM). We found no trials of self-help psychological interventions, psychological therapies or printed education. Future trials should include more diverse populations, address shared priorities, evaluate long-term outcomes and ensure patients are involved in trial design.

Topical anti-inflammatory treatments for eczema: network meta-analysis

BACKGROUND

Eczema (atopic dermatitis) is the most burdensome skin condition worldwide and cannot currently be prevented or cured. Topical anti-inflammatory treatments are used to control eczema symptoms, but there is uncertainty about the relative effectiveness and safety of different topical anti-inflammatory treatments.

OBJECTIVES

To compare and rank the efficacy and safety of topical anti-inflammatory treatments for people with eczema using a network meta-analysis.

SEARCH METHODS

We searched the Cochrane Skin Specialised Register, CENTRAL, MEDLINE, Embase and trial registries on 29 June 2023, and checked the reference lists of included studies.

SELECTION CRITERIA

We included within-participant or between-participant randomised controlled trials (RCTs) in people of any age with eczema of any severity, but excluded trials in clinically infected eczema, seborrhoeic eczema, contact eczema, or hand eczema. We included topical anti-inflammatory treatments used for at least one week, compared with another anti-inflammatory treatment, no treatment, or vehicle/placebo. Vehicle is a 'carrier system' for an active pharmaceutical substance, which may also be used on its own as an emollient for dry skin. We excluded trials of topical antibiotics used alone, complementary therapies, emollients used alone, phototherapy, wet wraps, and systemic treatments.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods. Primary outcomes were patient-reported eczema symptoms, clinician-reported eczema signs and investigator global assessment. Secondary outcomes were health-related quality of life, long-term control of eczema, withdrawal from treatment/study, and local adverse effects (application-site reactions, pigmentation changes and skin thinning/atrophy were identified as important concerns through patient and public involvement). We used CINeMA to quantify our confidence in the evidence for each outcome.

MAIN RESULTS

We included 291 studies involving 45,846 participants with the full spectrum of eczema severity, mainly conducted in high-income countries in secondary care settings. Most studies included adults, with only 31 studies limited to children aged < 12 years. Studies usually included male and female participants, multiple ethnic groups but predominantly white populations. Most studies were industry-funded (68%) or did not report their funding sources/details. Treatment duration and trial participation were a median of 21 and 28 days (ranging from 7 days to 5 years), respectively. Interventions used were topical corticosteroids (TCS) (172), topical calcineurin inhibitors (TCI) (134), phosphodiesterase-4 (PDE-4) inhibitors (55), janus kinase (JAK) inhibitors (30), aryl hydrocarbon receptor activators (10), or other topical agents (21). Comparators included vehicle (170) or other anti-inflammatory treatments. The risk of bias was high in 242 of the 272 (89.0%) trials contributing to data analyses, most commonly due to concerns about selective reporting. Network meta-analysis (NMA) was only possible for short-term outcomes. Patient-reported symptoms NMA of 40 trials (6482 participants) reporting patient-reported symptoms as a binary outcome ranked tacrolimus 0.1% (OR 6.27, 95% CI 1.19 to 32.98), potent TCS (OR 5.99, 95% CI 2.83 to 12.69), and ruxolitinib 1.5% (OR 5.64, 95% CI 1.26 to 25.25) as the most effective, all with low confidence. Mild TCS, roflumilast 0.15%, and crisaborole 2% were the least effective. Class-level sensitivity analysis found potent/very potent TCS had similar effectiveness to potent TCI and was more effective than mild TCI and PDE-4 inhibitors. NMA of 29 trials (3839 participants) reporting patient-reported symptoms as a continuous outcome ranked very potent TCS (SMD -1.99, 95% CI -3.25 to -0.73; low confidence) and tacrolimus 0.03% (SMD -1.57, 95% CI -2.42 to -0.72; moderate confidence) the highest. Direct information for tacrolimus 0.03% was based on one trial of 60 participants at high risk of bias. Roflumilast 0.15%, delgocitinib 0.25% or 0.5%, and tapinarof 1% were the least effective. Class-level sensitivity analysis found potent/very potent TCS had similar effectiveness to potent TCI and JAK inhibitors and mild/moderate TCS was less effective than mild TCI. A further 50 trials (9636 participants) reported patient-reported symptoms as a continuous outcome but could not be included in NMA. Clinician-reported signs NMA of 32 trials (4121 participants) reported clinician signs as a binary outcome and ranked potent TCS (OR 8.15, 95% CI 4.99, 13.57), tacrolimus 0.1% (OR 8.06, 95% CI 3.30, 19.67), ruxolitinib 1.5% (OR 7.72, 95% CI 4.92, 12.10), and delgocitinib 0.5% (OR 7.61, 95% CI 3.72, 15.58) as most effective, all with moderate confidence. Mild TCS, roflumilast 0.15%, crisaborole 2%, and tapinarof 1% were the least effective. Class-level sensitivity analysis found potent/very potent TCS more effective than potent TCI, mild TCI, JAK inhibitors, PDE-4 inhibitors; and mild TCS and PDE-4 inhibitors had similar effectiveness. NMA of 49 trials (5261 participants) reported clinician signs as a continuous outcome and ranked tacrolimus 0.03% (SMD -2.69, 95% CI -3.36, -2.02) and very potent TCS (SMD -1.87, 95% CI -2.69, -1.05) as most effective, both with moderate confidence; roflumilast 0.15%, difamilast 0.3% and tapinarof 1% were ranked as least effective. Direct information for tacrolimus 0.03% was based on one trial in 60 participants with a high risk of bias. For some sensitivity analyses, potent TCS, tacrolimus 0.1%, ruxolitinib 1.5%, delgocitinib 0.5% and delgocitinib 0.25% became some of the most effective treatments. Class-level analysis found potent/very potent TCS had similar effectiveness to potent TCI and JAK inhibitors, and moderate/mild TCS was more effective than mild TCI. A further 100 trials (22,814 participants) reported clinician signs as a continuous outcome but could not be included in NMA. Investigator Global Assessment NMA of 140 trials (23,383 participants) reported IGA as a binary outcome and ranked ruxolitinib 1.5% (OR 9.34, 95% CI 4.8, 18.18), delgocitinib 0.5% (OR 10.08, 95% CI 2.65, 38.37), delgocitinib 0.25% (OR 6.87, 95% CI 1.79, 26.33), very potent TCS (OR 8.34, 95% CI 4.73, 14.67), potent TCS (OR 5.00, 95% CI 3.80, 6.58), and tacrolimus 0.1% (OR 5.06, 95% CI 3.59, 7.13) as most effective, all with moderate confidence. Mild TCS, crisaborole 2%, pimecrolimus 1%, roflumilast 0.15%, difamilast 0.3% and 1%, and tacrolimus 0.03% were the least effective. In a sensitivity analysis of low risk of bias information (12 trials, 1639 participants), potent TCS, delgocitinib 0.5% and delgocitinib 0.25% were most effective, and pimecrolimus 1%, roflumilast 0.15%, difamilast 1% and difamilast 0.3% least effective. Class-level sensitivity analysis found potent/very potent TCS had similar effectiveness to potent TCI and JAK inhibitors and were more effective than PDE-4 inhibitors; mild/moderate TCS were less effective than potent TCI and had similar effectiveness to mild TCI. Longer-term outcomes over 6 to 12 months showed a possible increase in effectiveness for pimecrolimus 1% versus vehicle (4 trials, 2218 participants) in a pairwise meta-analysis, and greater treatment success with mild/moderate TCS than pimecrolimus 1% (based on 1 trial of 2045 participants). Local adverse effects NMA of 83 trials (18,992 participants, 2424 events) reporting application-site reactions ranked tacrolimus 0.1% (OR 2.2, 95% CI 1.53, 3.17; moderate confidence), crisaborole 2% (OR 2.12, 95% CI 1.18, 3.81; high confidence), tacrolimus 0.03% (OR 1.51, 95%CI 1.10, 2.09; low confidence), and pimecrolimus 1% (OR 1.44, 95% CI 1.01, 2.04; low confidence) as most likely to cause site reactions. Very potent, potent, moderate, and mild TCS were least likely to cause site reactions. NMA of eight trials (1786 participants, 3 events) reporting pigmentation changes found no evidence for increased pigmentation changes with TCS and crisaborole 2%, with low confidence for mild, moderate or potent TCS and moderate confidence for crisaborole 2%. NMA of 25 trials (3691 participants, 36 events) reporting skin thinning found no evidence for increased skin thinning with short-term (median 3 weeks, range 1-16 weeks) use of mild TCS (OR 0.72, 95% CI 0.12, 4.31), moderate TCS (OR 0.91, 95% CI 0.16, 5.33), potent TCS (OR 0.96, 95% CI 0.21, 4.43) or very potent TCS (OR 0.88, 95% CI 0.31, 2.49), all with low confidence. Longer-term outcomes over 6 to 60 months showed increased skin thinning with mild to potent TCS versus TCI (3 trials, 4069 participants, 6 events with TCS).

AUTHORS' CONCLUSIONS

Potent TCS, JAK inhibitors and tacrolimus 0.1% were consistently ranked as amongst the most effective topical anti-inflammatory treatments for eczema and PDE-4 inhibitors as amongst the least effective. Mild TCS and tapinarof 1% were ranked amongst the least effective treatments in three of five efficacy networks. TCI and crisaborole 2% were ranked most likely to cause local application-site reactions and TCS least likely. We found no evidence for increased skin thinning with short-term TCS but an increase with longer-term TCS.

Strategies for cessation of caffeine administration in preterm infants

BACKGROUND

Apnea and intermittent hypoxemia (IH) are common developmental disorders in infants born earlier than 37 weeks' gestation. Caffeine administration has been shown to lower the incidence of these disorders in preterm infants. Cessation of caffeine treatment is based on different post-menstrual ages (PMA) and resolution of symptoms. There is uncertainty about the best timing for caffeine discontinuation.

OBJECTIVES

To evaluate the effects of early versus late discontinuation of caffeine administration in preterm infants.

SEARCH METHODS

We searched CENTRAL, PubMed, Embase, and three trial registries in August 2023; we applied no date limits. We checked the references of included studies and related systematic reviews.

SELECTION CRITERIA

We included randomized controlled trials (RCTs) in preterm infants born earlier than 37 weeks' gestation, up to a PMA of 44 weeks and 0 days, who received caffeine for any indication for at least seven days. We compared three different strategies for caffeine cessation: 1. at different PMAs, 2. before or after five days without symptoms, and 3. at a predetermined PMA versus at the resolution of symptoms.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods. Primary outcomes were: restarting caffeine therapy, intubation within one week of treatment discontinuation, and the need for non-invasive respiratory support within one week of treatment discontinuation. Secondary outcomes were: number of episodes of apnea in the seven days after treatment discontinuation, number of infants with at least one episode of apnea in the seven days after treatment discontinuation, number of episodes of intermittent hypoxemia (IH) within seven days of treatment discontinuation, number of infants with at least one episode of IH in the seven days after of treatment discontinuation, all-cause mortality prior to hospital discharge, major neurodevelopmental disability, number of days of respiratory support after treatment discontinuation, duration of hospital stay, and cost of neonatal care. We used GRADE to assess the certainty of evidence for each outcome.

MAIN RESULTS

We included three RCTs (392 preterm infants). Discontinuation of caffeine at PMA less than 35 weeks' gestation versus PMA equal to or longer than 35 weeks' gestation This comparison included one single completed RCT with 98 premature infants with a gestational age between 25 + 0 and 32 + 0 weeks at birth. All infants had discontinued caffeine treatment for five days at randomization. The infants received either an oral loading dose of caffeine citrate (20 mg/kg) at randomization followed by oral maintenance dosage (6 mg/kg/day) until 40 weeks PMA, or usual care (controls), during which caffeine was stopped before 37 weeks PMA. Early cessation of caffeine administration in preterm infants at PMA less than 35 weeks' gestation may result in an increase in the number of IH episodes in the seven days after discontinuation of treatment, compared to prolonged caffeine treatment beyond 35 weeks' gestation (mean difference [MD] 4.80, 95% confidence interval [CI] 2.21 to 7.39; 1 RCT, 98 infants; low-certainty evidence). Early cessation may result in little to no difference in all-cause mortality prior to hospital discharge compared to late discontinuation after 35 weeks PMA (risk ratio [RR] not estimable; 98 infants; low-certainty evidence). No data were available for the following outcomes: restarting caffeine therapy, intubation within one week of treatment discontinuation, need for non-invasive respiratory support within one week of treatment discontinuation, number of episodes of apnea, number of infants with at least one episode of apnea in the seven days after discontinuation of treatment, or number of infants with at least one episode of IH in the seven days after discontinuation of treatment. Discontinuation based on PMA versus resolution of symptoms This comparison included two RCTs with a total of 294 preterm infants. Discontinuing caffeine at the resolution of symptoms compared to discontinuing treatment at a predetermined PMA may result in little to no difference in all-cause mortality prior to hospital discharge (RR 1.00, 95% CI 0.14 to 7.03; 2 studies, 294 participants; low-certainty evidence), or in the number of infants with at least one episode of apnea within the seven days after discontinuing treatment (RR 0.60, 95% CI 0.31 to 1.18; 2 studies; 294 infants; low-certainty evidence). Discontinuing caffeine based on the resolution of symptoms probably results in more infants with IH in the seven days after discontinuation of treatment (RR 0.38, 95% CI 0.20 to 0.75; 1 study; 174 participants; moderate-certainty evidence). No data were available for the following outcomes: restarting caffeine therapy, intubation within one week of treatment discontinuation, need for non-invasive respiratory support within one week of treatment discontinuation, or number of episodes of IH in the seven days after treatment discontinuation. Adverse effects In the Rhein 2014 study, five of the infants randomized to caffeine had the caffeine treatment discontinued at the discretion of the clinical team, because of tachycardia. The Pradhap 2023 study reported adverse events, including recurrence of apnea of prematurity (15% in the short and 13% in the regular course caffeine therapy group), varying severities of bronchopulmonary dysplasia, hyperglycemia, extrauterine growth restriction, retinopathy of prematurity requiring laser treatment, feeding intolerance, osteopenia, and tachycardia, with no significant differences between the groups. The Prakash 2021 study reported that adverse effects of caffeine therapy for apnea of prematurity included tachycardia, feeding intolerance, and potential neurodevelopmental impacts, though most were mild and transient. We identified three ongoing studies.

AUTHORS' CONCLUSIONS

There may be little or no difference in the incidence of all-cause mortality and apnea in infants who were randomized to later discontinuation of caffeine treatment. However, the number of infants with at least one episode of IH was probably reduced with later cessation. No data were found to evaluate the benefits and harms of later caffeine discontinuation for: restarting caffeine therapy, intubation within one week of treatment discontinuation, or need for non-invasive respiratory support within one week of treatment discontinuation. Further studies are needed to evaluate the short-term and long-term effects of different caffeine cessation strategies in premature infants.

Multisensory stimulation for promoting development and preventing morbidity in preterm infants

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the benefits and harms of multisensory stimulation compared to any single sensory intervention or standard care for physical and neurological development in preterm infants.

Shorter versus longer duration antibiotic regimens for treatment of culture-positive neonatal sepsis

OBJECTIVES

This is a protocol for a Cochrane Review (intervention). The objectives are as follows: To assess the safety and effectiveness of shorter versus longer duration antibiotic regimens for the treatment of culture-positive neonatal sepsis with or without meningitis.

Bone-modifying agents for reducing bone loss in women with early and locally advanced breast cancer: a network meta-analysis

BACKGROUND

Bisphosphonates and receptor activator of nuclear factor-kappa B ligand (RANKL)-inhibitors are amongst the bone-modifying agents used as supportive treatment in women with breast cancer who do not have bone metastases. These agents aim to reduce bone loss and the risk of fractures. Bisphosphonates have demonstrated survival benefits, particularly in postmenopausal women.

OBJECTIVES

To assess and compare the effects of different bone-modifying agents as supportive treatment to reduce bone mineral density loss and osteoporotic fractures in women with breast cancer without bone metastases and generate a ranking of treatment options using network meta-analyses (NMAs).

SEARCH METHODS

We identified studies by electronically searching CENTRAL, MEDLINE and Embase until January 2023. We searched various trial registries and screened abstracts of conference proceedings and reference lists of identified trials.

SELECTION CRITERIA

We included randomised controlled trials comparing different bisphosphonates and RANKL-inihibitors with each other or against no further treatment or placebo for women with breast cancer without bone metastases.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data and assessed the risk of bias of included studies and certainty of evidence using GRADE. Outcomes were bone mineral density, quality of life, overall fractures, overall survival and adverse events. We conducted NMAs and generated treatment rankings.

MAIN RESULTS

Forty-seven trials (35,163 participants) fulfilled our inclusion criteria; 34 trials (33,793 participants) could be considered in the NMA (8 different treatment options). Bone mineral density We estimated that the bone mineral density of participants with no treatment/placebo measured as total T-score was -1.34. Evidence from the NMA (9 trials; 1166 participants) suggests that treatment with ibandronate (T-score -0.77; MD 0.57, 95% CI -0.05 to 1.19) may slightly increase bone mineral density (low certainty) and treatment with zoledronic acid (T-score -0.45; MD 0.89, 95% CI 0.62 to 1.16) probably slightly increases bone mineral density compared to no treatment/placebo (moderate certainty). Risedronate (T-score -1.08; MD 0.26, 95% CI -0.32 to 0.84) may result in little to no difference compared to no treatment/placebo (low certainty). We are uncertain whether alendronate (T-score 2.36; MD 3.70, 95% CI -2.01 to 9.41) increases bone mineral density compared to no treatment/placebo (very low certainty). Quality of life No quantitative analyses could be performed for quality of life, as only three studies reported this outcome. All three studies showed only minimal differences between the respective interventions examined. Overall fracture rate We estimated that 70 of 1000 participants with no treatment/placebo had fractures. Evidence from the NMA (16 trials; 19,492 participants) indicates that treatment with clodronate or ibandronate (42 of 1000; RR 0.60, 95% CI 0.39 to 0.92; 40 of 1000; RR 0.57, 95% CI 0.38 to 0.86, respectively) decreases the number of fractures compared to no treatment/placebo (high certainty). Denosumab or zoledronic acid (51 of 1000; RR 0.73, 95% CI 0.52 to 1.01; 55 of 1000; RR 0.79, 95% CI 0.56 to 1.11, respectively) probably slightly decreases the number of fractures; and risedronate (39 of 1000; RR 0.56, 95% CI 0.15 to 2.16) probably decreases the number of fractures compared to no treatment/placebo (moderate certainty). Pamidronate (106 of 1000; RR 1.52, 95% CI 0.75 to 3.06) probably increases the number of fractures compared to no treatment/placebo (moderate certainty). Overall survival We estimated that 920 of 1000 participants with no treatment/placebo survived overall. Evidence from the NMA (17 trials; 30,991 participants) suggests that clodronate (924 of 1000; HR 0.95, 95% CI 0.77 to 1.17), denosumab (927 of 1000; HR 0.91, 95% CI 0.69 to 1.21), ibandronate (915 of 1000; HR 1.06, 95% CI 0.83 to 1.34) and zoledronic acid (925 of 1000; HR 0.93, 95% CI 0.76 to 1.14) may result in little to no difference regarding overall survival compared to no treatment/placebo (low certainty). Additionally, we are uncertain whether pamidronate (905 of 1000; HR 1.20, 95% CI 0.81 to 1.78) decreases overall survival compared to no treatment/placebo (very low certainty). Osteonecrosis of the jaw We estimated that 1 of 1000 participants with no treatment/placebo developed osteonecrosis of the jaw. Evidence from the NMA (12 trials; 23,527 participants) suggests that denosumab (25 of 1000; RR 24.70, 95% CI 9.56 to 63.83), ibandronate (6 of 1000; RR 5.77, 95% CI 2.04 to 16.35) and zoledronic acid (9 of 1000; RR 9.41, 95% CI 3.54 to 24.99) probably increases the occurrence of osteonecrosis of the jaw compared to no treatment/placebo (moderate certainty). Additionally, clodronate (3 of 1000; RR 2.65, 95% CI 0.83 to 8.50) may increase the occurrence of osteonecrosis of the jaw compared to no treatment/placebo (low certainty). Renal impairment We estimated that 14 of 1000 participants with no treatment/placebo developed renal impairment. Evidence from the NMA (12 trials; 22,469 participants) suggests that ibandronate (28 of 1000; RR 1.98, 95% CI 1.01 to 3.88) probably increases the occurrence of renal impairment compared to no treatment/placebo (moderate certainty). Zoledronic acid (21 of 1000; RR 1.49, 95% CI 0.87 to 2.58) probably increases the occurrence of renal impairment while clodronate (12 of 1000; RR 0.88, 95% CI 0.55 to 1.39) and denosumab (11 of 1000; RR 0.80, 95% CI 0.54 to 1.19) probably results in little to no difference regarding the occurrence of renal impairment compared to no treatment/placebo (moderate certainty).

AUTHORS' CONCLUSIONS

When considering bone-modifying agents for managing bone loss in women with early or locally advanced breast cancer, one has to balance between efficacy and safety. Our findings suggest that bisphosphonates (excluding alendronate and pamidronate) or denosumab compared to no treatment or placebo likely results in increased bone mineral density and reduced fracture rates. Our survival analysis that included pre and postmenopausal women showed little to no difference regarding overall survival. These treatments may lead to more adverse events. Therefore, forming an overall judgement of the best ranked bone-modifying agent is challenging. More head-to-head comparisons, especially comparing denosumab with any bisphosphonate, are needed to address gaps and validate the findings of this review.

Noise or sound management in the neonatal intensive care unit for preterm or very low birth weight infants

BACKGROUND

Infants in the neonatal intensive care unit (NICU) are subjected to different types of stress, including sounds of high intensity. The sound levels in NICUs often exceed the maximum acceptable level recommended by the American Academy of Pediatrics, which is 45 decibels (dB). Hearing impairment is diagnosed in 2% to 10% of preterm infants compared to only 0.1% of the general paediatric population. Bringing sound levels under 45 dB can be achieved by lowering the sound levels in an entire unit; by treating the infant in a section of a NICU, in a 'private' room, or in incubators in which the sound levels are controlled; or by reducing sound levels at the individual level using earmuffs or earplugs. By lowering sound levels, the resulting stress can be diminished, thereby promoting growth and reducing adverse neonatal outcomes. This review is an update of one originally published in 2015 and first updated in 2020.

OBJECTIVES

To determine the benefits and harms of sound reduction on the growth and long-term neurodevelopmental outcomes of neonates.

SEARCH METHODS

We used standard, extensive Cochrane search methods. On 21 and 22 August 2023, a Cochrane Information Specialist searched CENTRAL, PubMed, Embase, two other databases, two trials registers, and grey literature via Google Scholar and conference abstracts from Pediatric Academic Societies.

SELECTION CRITERIA

We included randomised controlled trials (RCTs) or quasi-RCTs in preterm infants (less than 32 weeks' postmenstrual age (PMA) or less than 1500 g birth weight) cared for in the resuscitation area, during transport, or once admitted to a NICU or stepdown unit. We specified three types of intervention: 1) intervention at the unit level (i.e. the entire neonatal department), 2) at the section or room level, or 3) at the individual level (e.g. hearing protection).

DATA COLLECTION AND ANALYSIS

We used the standardised review methods of Cochrane Neonatal to assess the risk of bias in the studies. We used the risk ratio (RR) and risk difference (RD), with their 95% confidence intervals (CIs), for dichotomous data. We used the mean difference (MD) for continuous data. Our primary outcome was major neurodevelopmental disability. We used GRADE to assess the certainty of the evidence.

MAIN RESULTS

We included one RCT, which enroled 34 newborn infants randomised to the use of silicone earplugs versus no earplugs for hearing protection. It was a single-centre study conducted at the University of Texas Medical School in Houston, Texas, USA. Earplugs were positioned at the time of randomisation and worn continuously until the infants were 35 weeks' postmenstrual age (PMA) or discharged (whichever came first). Newborns in the control group received standard care. The evidence is very uncertain about the effects of silicone earplugs on the following outcomes. • Cerebral palsy (RR 3.00, 95% CI 0.15 to 61.74)and Mental Developmental Index (MDI) (Bayley II) at 18 to 22 months' corrected age (MD 14.00, 95% CI 3.13 to 24.87); no other indicators of major neurodevelopmental disability were reported. • Normal auditory functioning at discharge (RR 1.65, 95% CI 0.93 to 2.94) • All-cause mortality during hospital stay (RR 2.07, 95% CI 0.64 to 6.70; RD 0.20, 95% CI -0.09 to 0.50) • Weight (kg) at 18 to 22 months' corrected age (MD 0.31, 95% CI -1.53 to 2.16) • Height (cm) at 18 to 22 months' corrected age (MD 2.70, 95% CI -3.13 to 8.53) • Days of assisted ventilation (MD -1.44, 95% CI -23.29 to 20.41) • Days of initial hospitalisation (MD 1.36, 95% CI -31.03 to 33.75) For all outcomes, we judged the certainty of evidence as very low. We identified one ongoing RCT that will compare the effects of reduced noise levels and cycled light on visual and neural development in preterm infants.

AUTHORS' CONCLUSIONS

No studies evaluated interventions to reduce sound levels below 45 dB across the whole neonatal unit or in a room within it. We found only one study that evaluated the benefits of sound reduction in the neonatal intensive care unit for hearing protection in preterm infants. The study compared the use of silicone earplugs versus no earplugs in newborns of very low birth weight (less than 1500 g). Considering the very small sample size, imprecise results, and high risk of attrition bias, the evidence based on this research is very uncertain and no conclusions can be drawn. As there is a lack of evidence to inform healthcare or policy decisions, large, well designed, well conducted, and fully reported RCTs that analyse different aspects of noise reduction in NICUs are needed. They should report both short- and long-term outcomes.

Long-Term Effect of Salt Substitution for Cardiovascular Outcomes : A Systematic Review and Meta-analysis

BACKGROUND

Salt substitution is a simple yet increasingly promising strategy to improve cardiovascular outcomes.

PURPOSE

To evaluate the long-term effects of salt substitution on cardiovascular outcomes.

DATA SOURCES

PubMed, EMBASE, Cochrane CENTRAL, and CINAHL searched from inception to 23 August 2023. Trial registries, citation analysis, and hand-search were also done.

STUDY SELECTION

Randomized controlled trials (RCTs) comparing provision of or advice to use a salt substitute with no intervention or use of regular salt among adults for 6 months or longer in total study duration.

DATA EXTRACTION

Two authors independently screened articles, extracted data, and assessed risk of bias. Primary outcomes include mortality, major cardiovascular events (MACE), and adverse events at 6 months or greater. Secondary and post hoc outcomes include blood pressure, cause-specific mortality, and urinary excretion at 6 months or greater. Random-effects meta-analyses were done and certainty of effect estimates were assessed using GRADE (Grading of Recommendations Assessment, Development and Evaluation).

DATA SYNTHESIS

Of the 16 included RCTs, 8 reported on primary outcomes. Most (n = 7 of 8) were done in China or Taiwan, 3 were done in residential facilities, and 7 included populations of older age (average 62 years) and/or with higher-than-average cardiovascular risk. In this population, salt substitute may reduce risk for all-cause mortality (6 RCTs; 27 710 participants; rate ratio [RR], 0.88 [95% CI, 0.82 to 0.93]; low certainty) and cardiovascular mortality (4 RCTs; 25 050 participants; RR, 0.83 [CI, 0.73 to 0.95]; low certainty). Salt substitute may result in a slight reduction in MACE (3 RCTs; 23 215 participants; RR, 0.85 [CI, 0.71 to 1.00]; very low certainty), with very low-certainty evidence of serious adverse events (6 RCTs; 27 995 participants; risk ratio, 1.04 [CI, 0.87 to 1.25]).

LIMITATIONS

The evidence base is dominated by a single, large RCT. Most RCTs were from China or Taiwan and involved participants with higher-than-average cardiovascular risk; therefore, generalizability to other populations is very limited.

CONCLUSION

Salt substitution may reduce all-cause or cardiovascular mortality, but the evidence for reducing cardiovascular events and for not increasing serious adverse events is uncertain, particularly for a Western population. The certainty of evidence is higher among populations at higher cardiovascular risk and/or following a Chinese diet.

PRIMARY FUNDING SOURCE

National Health and Medical Research Council. (PROSPERO: CRD42022327566).

Insights into the nutritional prevention of macular degeneration based on a comparative topic modeling approach

Topic modeling and text mining are subsets of natural language processing (NLP) with relevance for conducting meta-analysis (MA) and systematic review (SR). For evidence synthesis, the above NLP methods are conventionally used for topic-specific literature searches or extracting values from reports to automate essential phases of SR and MA. Instead, this work proposes a comparative topic modeling approach to analyze reports of contradictory results on the same general research question. Specifically, the objective is to identify topics exhibiting distinct associations with significant results for an outcome of interest by ranking them according to their proportional occurrence in (and consistency of distribution across) reports of significant effects. Macular degeneration (MD) is a disease that affects millions of people annually, causing vision loss. Augmenting evidence synthesis to provide insight into MD prevention is therefore of central interest in this article. The proposed method was tested on broad-scope studies addressing whether supplemental nutritional compounds significantly benefit macular degeneration. Six compounds were identified as having a particular association with reports of significant results for benefiting MD. Four of these were further supported in terms of effectiveness upon conducting a follow-up literature search for validation (omega-3 fatty acids, copper, zeaxanthin, and nitrates). The two not supported by the follow-up literature search (niacin and molybdenum) also had scores in the lowest range under the proposed scoring system. Results therefore suggest that the proposed method's score for a given topic may be a viable proxy for its degree of association with the outcome of interest, and can be helpful in the systematic search for potentially causal relationships. Further, the compounds identified by the proposed method were not simultaneously captured as salient topics by state-of-the-art topic models that leverage document and word embeddings (Top2Vec) and transformer models (BERTopic). These results underpin the proposed method's potential to add specificity in understanding effects from broad-scope reports, elucidate topics of interest for future research, and guide evidence synthesis in a scalable way. All of this is accomplished while yielding valuable and actionable insights into the prevention of MD.

Randomized controlled trials in nursing conducted by Latin American research teams: A scoping review

INTRODUCTION

Randomized controlled trials (RCTs) are the cornerstone of systematic reviews and other evidence synthesis. RCT identification remains challenging because of limitations in their indexation in major databases and potential language bias. Scientific production in Latin American nursing is steadily increasing, but little is known about its design or main features. We aimed to identify the extent of evidence from RCTs in nursing conducted by Latin American research teams and evaluate their main characteristics, including potential risk of bias.

DESIGN

Scoping review with risk of bias assessment.

METHODS

We conducted a scoping review including a comprehensive electronic search in five relevant databases. We completed a descriptive data analysis and a risk of bias assessment of eligible studies using Cochrane's guidance.

RESULTS

We identified 1784 references of which 47 were RCTs published in 40 journals. Twenty (42.6%) RCTs were published in journals in English. Chronic diseases were the most common health conditions studied (29.7%). Fifteen (31.9%) RCTs had a high risk of bias. Thirty (75%) journals were included in the Journal Citation Report (JCR) catalog and 5 (16.7%) were journals classified under nursing category. Twenty-one (52.5%) journals explicitly required CONSORT checklist recommendations for RCTs reporting.

CONCLUSION

Publication of RCTs in nursing by Latin American authors has increased. Most journals where RCTs are published are in English and not specific to nursing. Searches in journals of other disciplines may be necessary to facilitate identification of RCTs in nursing. CONSORT statements need to be actively promoted to facilitate rigorous methodology and reporting of RCTs.

CLINICAL RELEVANCE STATEMENT

This study highlights the need for an increased research focus on RCTs in nursing in Latin America, and the importance of enhancing the reporting quality of these studies to support evidence-based nursing practice.

Mobile health (m-health) smartphone interventions for adolescents and adults with overweight or obesity

BACKGROUND

Obesity is considered to be a risk factor for various diseases, and its incidence has tripled worldwide since 1975. In addition to potentially being at risk for adverse health outcomes, people with overweight or obesity are often stigmatised. Behaviour change interventions are increasingly delivered as mobile health (m-health) interventions, using smartphone apps and wearables. They are believed to support healthy behaviours at the individual level in a low-threshold manner.

OBJECTIVES

To assess the effects of integrated smartphone applications for adolescents and adults with overweight or obesity.

SEARCH METHODS

We searched CENTRAL, MEDLINE, PsycINFO, CINAHL, and LILACS, as well as the trials registers ClinicalTrials.gov and World Health Organization International Clinical Trials Registry Platform on 2 October 2023 (date of last search for all databases). We placed no restrictions on the language of publication.

SELECTION CRITERIA

Participants were adolescents and adults with overweight or obesity. Eligible interventions were integrated smartphone apps using at least two behaviour change techniques. The intervention could target physical activity, cardiorespiratory fitness, weight loss, healthy diet, or self-efficacy. Comparators included no or minimal intervention (NMI), a different smartphone app, personal coaching, or usual care. Eligible studies were randomised controlled trials of any duration with a follow-up of at least three months.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methodology and the RoB 2 tool. Important outcomes were physical activity, body mass index (BMI) and weight, health-related quality of life, self-efficacy, well-being, change in dietary behaviour, and adverse events. We focused on presenting studies with medium- (6 to < 12 months) and long-term (≥ 12 months) outcomes in our summary of findings table, following recommendations in the core outcome set for behavioural weight management interventions.

MAIN RESULTS

We included 18 studies with 2703 participants. Interventions lasted from 2 to 24 months. The mean BMI in adults ranged from 27 to 50, and the median BMI z-score in adolescents ranged from 2.2 to 2.5. Smartphone app versus no or minimal intervention Thirteen studies compared a smartphone app versus NMI in adults; no studies were available for adolescents. The comparator comprised minimal health advice, handouts, food diaries, smartphone apps unrelated to weight loss, and waiting list. Measures of physical activity: at 12 months' follow-up, a smartphone app compared to NMI probably reduces moderate to vigorous physical activity (MVPA) slightly (mean difference (MD) -28.9 min/week (95% confidence interval (CI) -85.9 to 28; 1 study, 650 participants; moderate-certainty evidence)). We are very uncertain about the results of estimated energy expenditure and cardiorespiratory fitness at eight months' follow-up. A smartphone app compared with NMI probably results in little to no difference in changes in total activity time at 12 months' follow-up and leisure time physical activity at 24 months' follow-up. Anthropometric measures: a smartphone app compared with NMI may reduce BMI (MD of BMI change -2.6 kg/m2, 95% CI -6 to 0.8; 2 studies, 146 participants; very low-certainty evidence) at six to eight months' follow-up, but the evidence is very uncertain. At 12 months' follow-up, a smartphone app probably resulted in little to no difference in BMI change (MD -0.1 kg/m2, 95% CI -0.4 to 0.3; 1 study; 650 participants; moderate-certainty evidence). A smartphone app compared with NMI may result in little to no difference in body weight change (MD -2.5 kg, 95% CI -6.8 to 1.7; 3 studies, 1044 participants; low-certainty evidence) at 12 months' follow-up. At 24 months' follow-up, a smartphone app probably resulted in little to no difference in body weight change (MD 0.7 kg, 95% CI -1.2 to 2.6; 1 study, 245 participants; moderate-certainty evidence). A smartphone app compared with NMI may result in little to no difference in self-efficacy for a physical activity score at eight months' follow-up, but the results are very uncertain. A smartphone app probably results in little to no difference in quality of life and well-being at 12 months (moderate-certainty evidence) and in little to no difference in various measures used to inform dietary behaviour at 12 and 24 months' follow-up. We are very uncertain about adverse events, which were only reported narratively in two studies (very low-certainty evidence). Smartphone app versus another smartphone app Two studies compared different versions of the same app in adults, showing no or minimal differences in outcomes. One study in adults compared two different apps (calorie counting versus ketogenic diet) and suggested a slight reduction in body weight at six months in favour of the ketogenic diet app. No studies were available for adolescents. Smartphone app versus personal coaching Only one study compared a smartphone app with personal coaching in adults, presenting data at three months. Two studies compared these interventions in adolescents. A smartphone app resulted in little to no difference in BMI z-score compared to personal coaching at six months' follow-up (MD 0, 95% CI -0.2 to 0.2; 1 study; 107 participants). Smartphone app versus usual care Only one study compared an app with usual care in adults but only reported data at three months on participant satisfaction. No studies were available for adolescents. We identified 34 ongoing studies.

AUTHORS' CONCLUSIONS

The available evidence is limited and does not demonstrate a clear benefit of smartphone applications as interventions for adolescents or adults with overweight or obesity. While the number of studies is growing, the evidence remains incomplete due to the high variability of the apps' features, content and components, which complicates direct comparisons and assessment of their effectiveness. Comparisons with either no or minimal intervention or personal coaching show minor effects, which are mostly not clinically significant. Minimal data for adolescents also warrants further research. Evidence is also scarce for low- and middle-income countries as well as for people with different socio-economic and cultural backgrounds. The 34 ongoing studies suggest sustained interest in the topic, with new evidence expected to emerge within the next two years. In practice, clinicians and healthcare practitioners should carefully consider the potential benefits, limitations, and evolving research when recommending smartphone apps to adolescents and adults with overweight or obesity.

Early developmental intervention programmes provided post hospital discharge to prevent motor and cognitive impairment in preterm infants

BACKGROUND

Infants born preterm are at increased risk of cognitive and motor impairments compared with infants born at term. Early developmental interventions for preterm infants are targeted at the infant or the parent-infant relationship, or both, and may focus on different aspects of early development. They aim to improve developmental outcomes for these infants, but the long-term benefits remain unclear. This is an update of a Cochrane review first published in 2007 and updated in 2012 and 2015.

OBJECTIVES

Primary objective To assess the effect of early developmental interventions compared with standard care in prevention of motor or cognitive impairment for preterm infants in infancy (zero to < three years), preschool age (three to < five years), and school age (five to < 18 years). Secondary objective To assess the effect of early developmental interventions compared with standard care on motor or cognitive impairment for subgroups of preterm infants, including groups based on gestational age, birthweight, brain injury, timing or focus of intervention and study quality.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, CINAHL, PsycINFO and trial registries in July 2023. We cross-referenced relevant literature, including identified trials and existing review articles.

SELECTION CRITERIA

Studies included randomised, quasi-randomised controlled trials (RCTs) or cluster-randomised trials of early developmental intervention programmes that began within the first 12 months of life for infants born before 37 weeks' gestational age (GA). Interventions could commence as an inpatient but had to include a post discharge component for inclusion in this review. Outcome measures were not prespecified, other than that they had to assess cognitive outcomes, motor outcomes or both. The control groups in the studies could receive standard care that would normally be provided.

DATA COLLECTION AND ANALYSIS

Data were extracted from the included studies regarding study and participant characteristics, timing and focus of interventions and cognitive and motor outcomes. Meta-analysis using RevMan was carried out to determine the effects of early developmental interventions at each age range: infancy (zero to < three years), preschool age (three to < five years) and school age (five to < 18 years) on cognitive and motor outcomes. Subgroup analyses focused on GA, birthweight, brain injury, time of commencement of the intervention, focus of the intervention and study quality. We used standard methodological procedures expected by Cochrane to collect data and evaluate bias. We used the GRADE approach to assess the certainty of evidence.

MAIN RESULTS

Forty-four studies met the inclusion criteria (5051 randomly assigned participants). There were 19 new studies identified in this update (600 participants) and a further 17 studies awaiting outcomes. Three previously included studies had new data. There was variability in the focus and intensity of the interventions, participant characteristics, and length of follow-up. All included studies were either single or multicentre trials and the number of participants varied from fewer than 20 to up to 915 in one study. The trials included in this review were mainly undertaken in middle- or high-income countries. The majority of studies commenced in the hospital, with fewer commencing once the infant was home. The focus of the intervention programmes for new included studies was increasingly targeted at both the infant and the parent-infant relationship. The intensity and dosages of interventions varied between studies, which is important when considering the applicability of any programme in a clinical setting. Meta-analysis demonstrated that early developmental intervention may improve cognitive outcomes in infancy (developmental quotient (DQ): standardised mean difference (SMD) 0.27 standard deviations (SDs), 95% confidence interval (CI) 0.15 to 0.40; P < 0.001; 25 studies; 3132 participants, low-certainty evidence), and improves cognitive outcomes at preschool age (intelligence quotient (IQ); SMD 0.39 SD, 95% CI 0.29 to 0.50; P < 0.001; 9 studies; 1524 participants, high-certainty evidence). However, early developmental intervention may not improve cognitive outcomes at school age (IQ: SMD 0.16 SD, 95% CI -0.06 to 0.38; P = 0.15; 6 studies; 1453 participants, low-certainty evidence). Heterogeneity between studies for cognitive outcomes in infancy and preschool age was moderate and at school age was substantial. Regarding motor function, meta-analysis of 23 studies showed that early developmental interventions may improve motor outcomes in infancy (motor scale DQ: SMD 0.12 SD, 95% CI 0.04 to 0.19; P = 0.003; 23 studies; 2737 participants, low-certainty evidence). At preschool age, the intervention probably did not improve motor outcomes (motor scale: SMD 0.08 SD, 95% CI -0.16 to 0.32; P = 0.53; 3 studies; 264 participants, moderate-certainty evidence). The evidence at school age for both continuous (motor scale: SMD -0.06 SD, 95% CI -0.31 to 0.18; P = 0.61; three studies; 265 participants, low-certainty evidence) and dichotomous outcome measures (low score on Movement Assessment Battery for Children (ABC) : RR 1.04, 95% CI 0.82 to 1.32; P = 0.74; 3 studies; 413 participants, low-certainty evidence) suggests that intervention may not improve motor outcome. The main source of bias was performance bias, where there was a lack of blinding of participants and personnel, which was unavoidable in this type of intervention study. Other biases in some studies included attrition bias where the outcome data were incomplete, and inadequate allocation concealment or selection bias. The GRADE assessment identified a lower certainty of evidence in the cognitive and motor outcomes at school age. Cognitive outcomes at preschool age demonstrated a high certainty due to more consistency and a larger treatment effect.

AUTHORS' CONCLUSIONS

Early developmental intervention programmes for preterm infants probably improve cognitive and motor outcomes during infancy (low-certainty evidence) while, at preschool age, intervention is shown to improve cognitive outcomes (high-certainty evidence). Considerable heterogeneity exists between studies due to variations in aspects of the intervention programmes, the population and outcome measures utilised. Further research is needed to determine which types of early developmental interventions are most effective in improving cognitive and motor outcomes, and in particular to discern whether there is a longer-term benefit from these programmes.