Publications by Year: 2026

BACKGROUND: Adolescents living with HIV(ALHIV) often experience lower retention in care and reduced viral suppression after transitioning to adult care. Differentiated care can enhance the uptake and utilisation of evidence-based transition interventions among populations with the greatest need. We aim to investigate the effectiveness, acceptability, feasibility, implementation, and costs of in-person and mHealth-based adolescent-friendly transition interventions.

METHODS: We are conducting a type 1 hybrid implementation-effectiveness design, with a cluster-randomised, stepped-wedge trial (SWT) to assess the effectiveness of in-person and mHealth-based adolescent-friendly transition interventions on retention in care and viral suppression among adolescents aged 15–19 years with perinatally-acquired HIV and low transition readiness. The SWT will be conducted in 16 clinics in urban (eThekwini) and rural (uMkhanyakude) KwaZulu-Natal over 24 months. Clinics will be randomly allocated to receive the intervention in period 1 (early) or period 2 (delayed). ALHIV, in the standard of care clinics, will be able to access the HIV prevention and treatment services delivered through the primary health clinics. We will use the RE-AIM (Reach, Effectiveness, Adoption, Implementation, and Maintenance) framework to guide our implementation outcomes. The primary outcomes for effectiveness are: (1) retention in care and (2) viral suppression measured in three cross-sectional surveys. We will measure retention in care as 80% of ART pharmacy refills on time, and 80% of scheduled clinic appointments attended, and we define viral suppression as < 200 copies/ml. We will compare the effectiveness and implementation outcomes of the in-person intervention with the mHealth intervention and compare outcomes between urban and rural clinics.

DISCUSSION: The findings of this trial will inform the expansion of differentiated care models for ALHIV to improve the transition process from paediatric to adult HIV care, optimised to support retention in care and viral suppression.

TRIAL REGISTRATION: ClinicalTrials.gov Identifier NCT06035445.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12913-026-14070-8.

Allosteric modulation is central to enzyme function and an attractive strategy for drug development. Protein Disulfide Isomerase (PDI), the prototypical thiol-isomerase, exemplifies this potential through its structural flexibility and involvement in neurodegeneration, cancer, and thromboinflammatory disorders such as sepsis, stroke, cancer-associated thrombosis, and antiphospholipid syndrome. PDI consists of four thioredoxin-like domains (a-b-b'-a'), with catalytic CGHC motifs in a and a' domains and a ligand-binding pocket in the b' domain. We previously reported that the b'-ligand bepristat 2a (Bep2a) inhibits PDI activity toward large macromolecular substrates while allosterically enhancing activity toward smaller physiological substrates such as GSSG and l-cystine. Here, we define the molecular, thermodynamic, and structural basis of this dual function. Bep2a features an indole ring with five substituents (R1-R5). Using mutagenesis and HDX-MS, we mapped the complex topology, identified five residues (F249, H256, I301, F304, I318) involved in binding, and uncovered a ligand-induced rearrangement of the left helix that acts as a dynamic gate controlling pocket accessibility, a previously unrecognized regulatory mechanism. AI-informed modeling, SAR analysis, and smFRET revealed that Bep2a's indole core binds perpendicularly in the pocket, with the R1 hydroxyl forming a critical hydrogen bond with H256, which is essential for binding but not for allosteric activation. Conversely, the R4 amine projects outward, serving as a key allosteric site that engages the catalytic domains and promotes PDI compaction. These findings uncover fundamental principles of PDI allosteric regulation and provide a blueprint for optimizing existing ligands and designing new ones with defined functional outcomes.

INTRODUCTION: Per- and polyfluoroalkyl substances (PFAS) are a group of synthetic chemicals used in consumer products and industrial applications. Higher prenatal PFAS levels have been linked to adverse pregnancy outcomes, yet underlying biological mechanisms are not well understood, although disruptions in inflammatory processes are suspected.

OBJECTIVES: Examine associations between prenatal PFAS exposure and maternal bioactive lipid concentrations during pregnancy.

METHODS: We included 436 participants from the racially and socioeconomically diverse LIFECODES cohort in Boston, Massachusetts. Maternal plasma samples collected in early pregnancy [median gestation = 10 weeks] were analyzed for nine PFAS, while a panel of 55 bioactive lipids, including polyunsaturated fatty acid precursors and their oxylipin derivatives, were assessed in samples collected at up to two gestational time points (median gestation = 9.6 and 26 weeks). We examined associations between individual PFAS and repeated measures of bioactive lipids using generalized estimating equation models, adjusting for maternal demographic, socioeconomic, and study-related factors. We additionally conducted study visit- and fetal sex-stratified analyses.

RESULTS: Higher concentrations of certain PFAS were associated with significant changes in bioactive lipid levels, with direction and magnitude varying by specific PFAS, bioactive lipid, and gestational timing. Several PFAS were inversely associated with prostaglandins (PGs), hydroxyeicosatetraenoic acids (HETEs), and oxoeicosatetraenoic acids (oxoETEs) across multiple enzymatic pathways. In contrast, we observed positive associations of several PFAS with leukotriene B4 (LTB4), with an interquartile range (IQR) increase in perfluorooctanoic acid (PFOA) leading to a 14.35% (95% CI: 5.03, 24.5) increase in LTB4 levels. Stratified analysis showed generally stronger associations at the earlier gestational time point than at the later time point, while trends by fetal sex were inconsistent.

CONCLUSIONS: Prenatal PFAS exposure was associated with alterations in maternal bioactive lipid levels, particularly derivatives of arachidonic acid. These shifts suggest that disruption of arachidonic acid metabolism may be an important pathway through which PFAS may exert their biological effects, warranting further mechanistic investigation.

This study investigated age differences in the precise knowledge and imprecise knowledge, or awareness, of multiple moving visual objects, measured by Multiple Identity Tracking (MIT) and Multiple Object Awareness (MOA) capacities, respectively, in a multiple object tracking task. Experiment 1 demonstrated significant decline of both capacities in older observers (65-80 years) compared to younger observers (18-44 years). Experiment 2 showed that age-related declines in MIT and MOA were linear across the adult lifespan (18-76 years). Additionally, we used computational models to test whether age effects could be explained by one common signal-strength factor (d') or by a dual-process model with an additional recollection parameter (R). Our results indicate that a detailed, recollection-based object-location representation (R) only plays a small role in tracking many objects and this factor does not vary with observers' age. For most observers, a single signal-strength parameter (d) explained behaviour best, and this parameter significantly declined with observers' age. This suggests that reduced sensitivity likely impairs older adults' ability to discriminate and clearly represent visual objects, resulting in both lower MIT and MOA capacities.

Major Depressive Disorder (MDD) is one of the most prevalent psychological disorders and frequently co-occurs with alcohol use disorders, increasing the risk of functional impairment. Monitoring alcohol use during depression treatment is therefore critical for early intervention. Passively collected data via devices like smartphones and smartwatches, offers a low-burden method for monitoring behavior in real time. This study investigated whether deep learning models trained on passively collected data (i.e., accelerometer, heart rate, respiratory rate, screen usage, and GPS data) could detect and predict alcohol use in individuals with MDD. Data were collected from 300 clinically depressed individuals who were enrolled in the Tracking Depression Study, a 90-day longitudinal study. Participants self-reported their alcohol use every week by completing the Timeline FollowBack. We trained models to predict same-day and next-day alcohol use. To validate these models, we split the data by participant, so that predictions were made on individuals who were not included in the training set. The models achieved moderate performance (mean AUC = 0.67 for both prediction tasks) when capturing both interindividual (between-person) and intraindividual (within-person) variability. Similar performances were observed when evaluating the model exclusively on predicting intraindividual variability (AUCs = 0.69 same-day, 0.68 next-day). However, model performance remained comparable to a baseline using only the day of week as predictor. These findings suggest that much of the predictive signal derives from temporal patterns. This indicates that interventions aligned with such temporal cues may already be effective, and that the added value of our model appears limited.

BACKGROUND AND AIM: Higher dietary flavonoid intake has been associated with lower risks of mortality and major chronic disease, yet its relationship with psychological well-being (PWB), a key contributor to health and quality of life, remains unclear. This study aimed to investigate bidirectional associations between dietary flavonoid intake and two PWB facets: happiness (positive emotional state) and optimism (generalized expectation of positive outcomes). Specifically, we examined whether (1) overall flavonoid-rich dietary patterns (flavodiet score), (2) intake of specific flavonoid-rich foods, and (3) total flavonoid and subclass intakes were each associated with happiness and optimism over time.

METHODS: Data were drawn from the Nurses' Health Study to form two analytical samples. Flavonoid intake measured in 1990 (n = 44,659) was examined in relation to sustained happiness (1992-2000) while intake in 2002 (n = 36,723) was analysed in relation to sustained optimism (2004-2012). Secondary analyses assessed whether higher baseline levels of each PWB facet were associated with sustained higher flavonoid intake, over up to 18 years. Associations were assessed using generalized estimating equations, adjusting for potential confounders.

RESULTS: Higher flavodiet scores were associated with a 3-6 % higher likelihood of sustained happiness [RRQ4vsQ1 (95 % CI): 1.03 (1.02-1.05)] and optimism [RRQ4vsQ1 (95 % CI): 1.06 (1.01-1.11)]. Specific flavonoid-rich foods (strawberries, apples, oranges, grapefruit, blueberries) were associated with a 3-16 % greater likelihood of sustained PWB, across the two facets. Similarly, total flavonoid and subclass intakes were associated with a 2-18 % greater likelihood of sustained PWB. Women with higher baseline levels of happiness or optimism were also more likely to sustain a higher flavonoid intake.

CONCLUSIONS: Consuming ∼3 servings/day of flavonoid-rich foods is associated with sustained PWB, and higher baseline PWB is associated with sustained higher flavonoid intake over up to 18 years. This bidirectional relationship suggests that integrated interventions targeting both diet and well-being may help promote long-term health and reduce chronic disease risk.

BACKGROUND: Surgeon volume and experience may impact patient-reported outcome measures (PROMs) and costs in total knee arthroplasty. However, whether the same relationship exists in unicompartmental knee arthroplasty (UKA) is unclear. We investigated whether surgeon volume and experience drove variations in PROMs or time-driven activity-based costing (TDABC) in UKA.

METHODS: We sourced data from a prospectively maintained multi-institutional arthroplasty registry. Patients completed the Knee Injury and Osteoarthritis Outcome Score - physical function short-form (KOOS-PS) with thresholds for minimal clinically important difference (MCID) and patient acceptable symptom state (PASS). A stratum-specific likelihood ratio analysis was used to categorize surgeons into volume and experience levels: low volume (annual volume less than 16), mid volume (annual volume 16 to 40), and high volume (annual volume greater than 40). The stratum-specific likelihood ratio analysis did not identify meaningful thresholds for surgeon experience. Because not all registry centers perform TDABC, our PROM analysis investigated 794 UKAs performed by 32 surgeons, while our TDABC subanalysis investigated 416 UKAs performed by 12 surgeons. Chi-square tests and one-way analyses of variance compared MCID, PASS, and costs between groups.

RESULTS: Low-volume UKA surgeons were associated with prolonged operating room times (low volume: 110 versus high volume: 73 minutes; P < 0.001) and lengths of stay (low volume: 1.2 versus high volume: 0.6 days; P < 0.001). Both MCID and PASS achievements were similar across volume levels (P = 0.80 and 0.84, respectively). However, low-volume surgeons were associated with increased costs (low volume: 744 versus mid volume: 662 and high volume: 662 cost units; P < 0.001).

CONCLUSIONS: Low-volume UKA surgeons had increased costs. However, MCID and PASS achievements were similar for UKA surgeons of all volume levels. There were no meaningful thresholds for surgeon experience. These results may reassure surgeons of all experience levels to pursue UKAs, while remaining aware of the positive influence of volume on cost-effectiveness.

In January 2021, the Centers for Medicare and Medicaid Services (CMS) implemented changes that increased work relative-value units (wRVUs) for outpatient evaluation and management (E/M) services, potentially benefiting cognitive specialties such as pediatric dermatology. This cross-sectional study analyzed 230,524 dermatology outpatient encounters across pediatric, general adult, and micrographic surgery and dermatologic oncology (MSDO) providers at a large academic health system in 2019 and 2021 to assess the impact of these changes. We found that mean wRVUs per encounter increased in pediatric and adult dermatology and decreased for micrographic surgery after the CMS revision (pediatric: 1.3-1.9, p < 0.01; adult: 1.8-2.2, p < 0.01; MSDO: 12.9-12.6, p = 0.048). Both pediatric and adult dermatologists shifted toward billing higher-level E/M codes after the 2021 reforms. These findings suggest that the 2021 CMS reforms modestly narrowed, but did not eliminate-longstanding reimbursement disparities between cognitive and procedural dermatology practices.