Transgender and gender-diverse individuals often face significant barriers to accessing gender-affirming care, such as hormones and/or surgery, leading to poorer mental health, lower quality of life, and higher rates of substance use and suicidal ideation. Vaginoplasty, the most commonly sought genital gender-affirming surgery (GGAS), is desired by over half of all trans women but has been performed in only a minority. This is due largely to limited surgeon availability and long wait times. Peer support has been shown to improve health outcomes and reduce stigma in marginalised populations, including trans communities, but has never been studied for efficacy during the perioperative period of GGAS. Building on priorities identified by multi-stakeholder engagement from the Transgender & Non-Binary Surgery Allied Research Collective, the Support for Transgender and Nonbinary Individuals Seeking Vaginoplasty (STRIVE) study aims to evaluate the efficacy of a centralised peer support and education intervention for patients seeking vaginoplasty, addressing a critical gap in perioperative care.

The STRIVE Study is a pragmatic, multi-site randomised controlled trial enrolling trans adults seeking full depth vaginoplasty. Participants are randomised to one of two arms; enhanced usual care, or a facilitated group intervention. The primary outcome is coping self-efficacy at 6 months, with a secondary outcome of surgical readiness. Primary analysis uses an intention-to-treat approach with linear mixed effects modelling, adjusting for selected baseline values and site. The feasibility evaluation data collected via qualitative interviews will be analysed thematically.

Approvals were granted by the primary site’s Institutional Review Board on 10 May 2024 (STUDY00026957). The trial was registered on 24 May 2024. Results will be published in open access journals and made available to community members in plain language formats.

NCT06436560.

To characterise long-term trajectory of recovery in individuals with long covid.

Prospective cohort.

Single-centre, specialist post-COVID service (London, UK).

Individuals aged ≥18 years with long covid (hospitalised and non-hospitalised) from April 2020 to March 2024.

Routine, prospectively collected data on symptoms, quality of life (including Fatigue Assessment Scale (FAS) and EuroQol 5 Dimensions (EQ-5D), return to work status and healthcare utilisation (investigations, outpatient and emergency attendances). The primary outcome was recovery by self-reported >75% of ‘best health’ (EQ-5D Visual Analogue Scale) and was assessed using Cox proportional hazards regression models over 4 years. Linked National Health Service England registry data provided secondary care healthcare utilisation and expenditure.

We included 3590 individuals (63.3% female, 73.5% non-hospitalised, median age 50.0 years, 71.9% with ≥2 doses of COVID-19 vaccination), who were followed up for a median of 136 (0–346) days since first assessment and 502 (251–825) days since symptom onset. At first assessment, 33.2% of employed individuals were unable to work. Dominant symptoms were fatigue (78.7%), breathlessness (68.1%) and brain fog (53.5%). 33.4% of individuals recovered to >75% of best health prior to clinic discharge (recovery occurred median 202 (94–468) days from symptom onset). Vaccinated individuals were more likely to recover faster (pre: HR 2.93 (2.00–4.28) and post: HR 1.34 (1.05–1.71) COVID-19 infection), whereas recovery hazard was inversely associated with FAS (HR 0.37 (0.33–0.42)), myalgia (HR 0.59 (0.45–0.76)) and dysautonomic symptoms (HR 0.46 (0.34–0.62)). There was high secondary care healthcare utilisation (both emergency and outpatient care). Annual inpatient and outpatient expenditure was significantly lower in hospitalised individuals while under the service. When compared with the prereferral period, emergency department attendances were reduced in non-hospitalised patients with long covid, but outpatient costs increased.

In the largest long covid cohort from a single specialist post-COVID service to date, only one-third of individuals under follow-up achieved satisfactory recovery. Fatigue severity and COVID-19 vaccination at presentation, even after initial COVID-19 infection, was associated with long covid recovery. Ongoing service provision for this and other post-viral conditions is necessary to support care, progress treatment options and provide capacity for future pandemic preparedness. Research and clinical services should emphasise these factors as the strongest predictors of non-recovery.

To assess the capability of a convolutional neural network trained by transfer learning on anterior segment optical coherence tomography (AS-OCT) images, Placido-disk corneal topography images and external photographs to predict age and biological sex.

Development of a deep learning model trained on retrospectively collected data using transfer learning.

A multicentre secondary care public health trust based in London.

We included 557,468 scans from 40,592 eyes of 20,542 patients. Data were extracted from all patients who underwent MS-39 imaging within our trust from October 2020 to March 2023.

Primary outcome measures for biological sex classification included accuracy, precision, recall, F1-score and area under the receiver operating curve (ROC-AUC). Primary outcome measures for age prediction were Pearson correlation coefficients (r), coefficients of determination (R²) and the mean absolute error (MAE) to evaluate the predictive performance. The secondary outcome was to visualise and interpret the model’s decision-making process through the construction of saliency maps.

For age prediction, the MAEs for the Placido, AS-OCT and external photograph models were 5.2, 5.1 and 6.2 years, respectively. For gender classification, the same models achieved ROC-AUCs of 0.88, 0.73 and 0.81, respectively. No difference in performance was found in the analysis of corneas with pathological topography. The saliency maps highlighted the peri-limbal cornea for age prediction and the central cornea for gender discrimination.

Our study demonstrates that deep learning models can extract age and gender information from anterior segment images. These findings support the concept that the anterior segment, like the retina, encodes important biological information. Future research should explore whether these models can predict specific systemic conditions.

Intrathoracic cancers, such as lung cancer, mesothelioma and thymoma, represent diagnostic challenges in primary care. We aimed to summarise evidence on the performance of imaging techniques that could aid the detection of intrathoracic cancers in low prevalence settings.

Systematic review and quality appraisal using Quality Assessment of Diagnostic Accuracy Studies-2 and Grading of Recommendations Assessment, Development and Evaluation.

MEDLINE, Embase and Web of Science were searched with a predesigned search strategy for articles from January 2000 to January 2024.

We included studies relevant for primary care, where participants were suspected of having intrathoracic cancer and reported on at least one diagnostic performance measure. We excluded studies where the cancer diagnosis was already established. Data extraction and synthesis screening were conducted independently by two reviewers. Data extraction and quality appraisal were conducted by one reviewer and checked by a second reviewer.

Out of 30 539 records identified by the database searches, 13 studies were included. There was heterogeneity in the types of cancers, populations included and reported diagnosis pathways for suspected cancers. Imaging modalities investigated included chest X-ray (three studies), computer tomography (CT, six studies), magnetic resonance imaging (two studies), positron emission tomography CT (two studies), ultrasound (two studies) and scintigraphy (one study). Chest X-ray sensitivity reported for lung cancer ranged from 33.3% to 75.9%, with specificity ranging from 83.2% to 95.5%. For CT, reported sensitivity varied from 58% for pleural malignancy to 100% for lung cancer. One study investigating an artificial intelligence tool to detect lung cancer found poor detection performance in a real-world patient cohort.

We found a limited number of studies reporting on the diagnostic performance of usual imaging techniques when used in unselected primary care settings for the diagnosis of intrathoracic cancer in symptomatic patients. There is a need for more studies evaluating such techniques in the general population presenting in primary care, where the prevalence is relatively low. A better understanding of the performance could lead to better detection strategies for intrathoracic cancers in primary care. Intrathoracic cancers, such as lung cancer, mesothelioma and thymoma, represent diagnostic challenges in primary care. We aimed to summarise evidence on the performance of imaging techniques that could aid the detection of intrathoracic cancers in low prevalence settings.