Nevin Manimala

Clin Obes. 2026 Apr;16(2):e70077. doi: 10.1111/cob.70077.

ABSTRACT

Metabolic and bariatric surgery (MBS) is an effective treatment for severe obesity but remains substantially underutilized. Limited data exist on outpatient counselling patterns preceding surgical uptake. Using electronic health record data, we conducted a 10-year retrospective cohort study of adults with body mass index ≥ 40 kg/m². The primary outcome was documentation of the MBS discussion. Secondary outcomes included MBS receipt and demographic factors associated with documented discussion. Among 60 574 eligible patients, only 7.6% had documented MBS discussion. Overall, 1.2% underwent MBS. Surgery occurred in 12.1% of patients with documented discussion compared with 0.3% without discussion. Patients with documented discussion were younger (median, 42 vs. 47 years), had higher BMI (median, 46.8 vs. 42.0 kg/m²), and were more often female. Eligible women were more likely than men to have documented discussion (8.7% vs. 5.5%). Black patients had higher discussion rates than White patients, despite known downstream disparities in MBS utilization. Documented MBS discussion is rare but represents a critical inflection point in surgical uptake. Demographic differences in counselling suggest clinician- and system-level factors influence access before referral, highlighting outpatient counselling as a key target to improve equitable MBS utilization.

PMID:41871890 | DOI:10.1111/cob.70077

Open Heart. 2026 Mar 23;13(1):e003969. doi: 10.1136/openhrt-2026-003969.

ABSTRACT

BACKGROUND: Transcatheter aortic valve replacement (TAVR) has already been recommended for some high-risk patients with aortic valve regurgitation, but there is still a lack of evidence regarding its early-term and medium-term safety and effectiveness compared with surgical aortic valve replacement (SAVR).

METHODS: This retrospective study included patients who underwent bioprosthetic aortic valve replacement for severe aortic regurgitation (AR) at a single centre between January 2018 and December 2023. All patients in the TAVR group received the J-Valve system via transapical (TA) approach. Propensity score matching (PSM) was used to balance the groups. The primary endpoint was 2-year all-cause mortality. Secondary endpoints included other clinical events, left ventricular (LV) function recovery and prosthesis haemodynamics, assessed by transthoracic echocardiography.

RESULTS: A total of 369 patients (median age 68 years, 26.6% female) were enrolled. Of these, 256 underwent TA-TAVR and 113 underwent SAVR. After 1:1 PSM, 76 matched pairs were included. There were no statistical differences between the groups in all-cause mortality, cardiovascular mortality, stroke, heart failure rehospitalisation, permanent pacemaker implantation or moderate to severe paravalvular leakage at 30 days or 2 years. Before PSM, left ventricular ejection fraction (LVEF) improved in the TAVR group (57% (IQR: 45-63%) vs 61% (IQR: 55-65%), p<0.001), with no significant change in the SAVR group (61% (IQR: 55-65%) vs 62% (IQR: 59-66%), p>0.05). After PSM, LVEF improvement was comparable between groups (+4.0% (IQR: -1.5 to 10.0) vs +2.0% (IQR: -3.0 to 9.5), p=0.430). Haemodynamics was superior in the TAVR group (p<0.001), while regression of LV dimensions was greater in the SAVR group.

CONCLUSION: In patients with severe AR, using the J-Valve for TA-TAVR showed comparable outcomes to SAVR regarding mortality and other clinical events. TAVR provided superior valve haemodynamics and was an effective treatment that significantly improved LV function, especially in high-risk patients.

PMID:41871886 | DOI:10.1136/openhrt-2026-003969

BMJ Ment Health. 2026 Mar 23;29(1):e302066. doi: 10.1136/bmjment-2025-302066.

ABSTRACT

BACKGROUND: Postpartum depression (PPD) remains vastly underdiagnosed, and its clinical heterogeneity is not well understood. Diagnosis codes in electronic health records (EHRs) alone may not identify all PPD cases, highlighting a need for novel detection approaches.

OBJECTIVE: To develop a transformer-based natural language processing (NLP) method to identify patients with PPD from clinical notes in EHRs and to examine demographic and clinical heterogeneity among identified cases.

METHODS: Clinical notes from 64 426 patients who gave birth between 2010 and 2023 at a major US academic medical centre were used to develop and evaluate the NLP method. By augmenting the NLP output with International Classification of Diseases (ICD-9/10) diagnosis codes, three subgroups of individuals with PPD were identified: patients identified by ICD only (PPD-ICD), NLP only (PPD-NLP) and both ICD and NLP (PPD-BOTH). Demographics, mental health and substance use disorders (SUDs), antidepressant treatment, behavioural therapy and healthcare utilisation were compared across PPD subgroups and a non-PPD control group. Longitudinal associations of depression and anxiety were also examined.

FINDINGS: The NLP method identified an additional 29.6% of patients whose clinical notes indicated symptoms suggestive of PPD but who lacked an ICD diagnosis. Significant variation was observed among PPD subgroups in comorbid psychiatric disorders, SUDs, treatment patterns and healthcare utilisation. During the 24 months post-delivery, the PPD-BOTH subgroup exhibited the highest rates of anxiety disorder diagnoses (vs PPD-ICD: OR 1.69, 95% CI 1.49 to 1.93; vs PPD-NLP: OR 4.46, 95% CI 3.82 to 5.22), antidepressant prescriptions (vs PPD-ICD: OR 1.95, 95% CI 1.71 to 2.22; vs PPD-NLP: OR 5.98, 95% CI 5.11 to 7.01) and mental health outpatient visits (vs PPD-ICD: OR 1.45, 95% CI 1.24 to 1.7; vs PPD-NLP: OR 4.94, 95% CI 3.9 to 6.31), suggesting higher symptom severity (all p<0.001). Comorbid depression and anxiety diagnoses were most prevalent during the postpartum period and declined over time.

CONCLUSIONS: Augmenting NLP-based identification with ICD codes yielded more individuals with distinct demographic and clinical profiles, demonstrating the method’s ability to improve case detection and characterise heterogeneity.

CLINICAL IMPLICATIONS: Given that PPD is underdiagnosed and undertreated, this novel approach demonstrates further potential for NLP in healthcare settings to capture more cases, enabling earlier and more personalised interventions that reach patients who may otherwise be overlooked.

PMID:41871883 | DOI:10.1136/bmjment-2025-302066

BMJ Glob Health. 2026 Mar 23;11(3):e021659. doi: 10.1136/bmjgh-2025-021659.

ABSTRACT

BACKGROUND: Injury burden is high in low-income and middle-income countries (LMICs). Delays in accessing definitive care after injury beyond the ‘golden’ hour or 2 hours worsen outcomes. We examined delays in accessing definitive healthcare after injury and whether their magnitude and associations differ across four diverse LMICs: Ghana, Pakistan, Rwanda and South Africa.

METHODS: Across 19 hospitals providing definitive care for injuries in urban or rural settings, we enrolled patients with moderate to severe injuries who were hospitalised for at least 12 hours. The time between injury and admission for definitive care and perceived reasons for delays in seeking and reaching care were captured. The association between more than 1-hour delay to reaching definitive care and age, sex, education, wealth, injury mechanism or severity, prior healthcare encounters, ambulance transport, the hospital type and catchment area was evaluated in a multivariable model. Patients’ perceived reasons for delay in seeking and reaching care were described. Findings were compared between countries.

RESULT: Data on delays were available for 8331 patients, of whom 57.3% experienced delays exceeding 1 hour. Prior healthcare encounter before definitive care showed the strongest association with delay (OR: 8.44, 95% CI 7.41 to 9.60). Delays were associated with older age, less education and wealth, greater injury severity, urban (vs rural) catchment area, ambulance transport, injury mechanism due to falls or fire (vs road traffic collision) and tertiary (vs secondary) hospital admission in the adjusted model. Ghana and Rwanda showed the lowest and highest odds of delays compared with South Africa, respectively. Only 18.8% of patients perceived being delayed, most citing unawareness of urgency and ambulance unavailability as reasons.

CONCLUSIONS: Most injured patients do not arrive at definitive care within the critical golden hour, with delays inequitably affecting the population. Improvements in pathways to care are needed to reduce delays across healthcare systems.

PMID:41871872 | DOI:10.1136/bmjgh-2025-021659

BMJ Health Care Inform. 2026 Mar 23;33(1):e101877. doi: 10.1136/bmjhci-2025-101877.

ABSTRACT

We present BODHI (Balanced, Open-minded, Diagnostic, Humble, and Inquisitive), an engineering framework for curiosity driven and humble clinical decision support artificial intelligence (AI) systems. Despite growing capabilities, large language models (LLMs) often express inappropriate confidence, conflating statistical pattern recognition with genuine medical understanding. BODHI addresses this through a dual reflective architecture that: (1) decomposes epistemic uncertainty into task specific dimensions, and (2) constrains model responses using virtue based stance rules derived from a Virtue Activation Matrix. We validate the framework through controlled evaluation on 200 clinical vignettes from HealthBench Hard, assessing GPT-4o-mini and GPT-4.1-mini across 5 random seeds (2000 total observations). Statistical analysis included bootstrap resampling, paired t tests, and effect size computation. BODHI improved overall clinical response quality (GPT-4.1-mini: +16.6 pp, p<0.0001, Cohen’s d=11.56; GPT-4o-mini: +2.2 pp, p<0.0001, Cohen’s d=1.56) and achieved very large effect sizes on curiosity (context seeking rate: Cohen’s d=16.38 and 19.54) and humility (hedging: d=5.80 for GPT-4.1-mini) metrics. Crucially, 97.3% of GPT-4.1-mini responses and 73.5% of GPT-4o-mini responses included appropriate clarifying questions, compared with 7.8% and 0.0% at baseline, demonstrating the framework’s effectiveness in eliciting information gathering behaviour. Findings suggest LLMs can be reliably constrained to operate within epistemic boundaries when provided with structured uncertainty decomposition and virtue aligned response rules, offering a pathway towards safer clinical AI deployment.

PMID:41871866 | DOI:10.1136/bmjhci-2025-101877

Psychol Rep. 2026 Mar 23:332941261436723. doi: 10.1177/00332941261436723. Online ahead of print.

ABSTRACT

Survivors of intimate partner violence (IPV) experience significant psychological consequences including high rates of suicidal thoughts and behaviors. A history of child maltreatment (CM) is also prevalent among IPV survivors and has been identified as a significant risk factor for suicide. Posttraumatic stress disorder (PTSD) symptoms have been proposed as a mechanism by which CM leads to suicide risk; however, this association has yet to be evaluated in IPV survivors. In the current study, we tested whether CM was associated with suicide risk among IPV survivors and whether this association was statistically explained by PTSD symptoms. A total of 122 adult survivors of IPV completed a survey containing measures of CM, IPV victimization experiences, PTSD symptoms, and suicide risk. Five mediation analyses examined direct and indirect effects of each type of CM (i.e., physical abuse, emotional abuse, sexual abuse, physical neglect, and emotional neglect) on suicide risk. Across all models, IPV victimization was associated with greater PTSD symptoms. All abuse subtypes of CM were associated with greater PTSD symptoms while the neglect subtypes of CM were not associated with PTSD symptoms. There was no direct effect of any type of CM on suicide risk; however, we found that greater experiences of childhood emotional abuse, physical abuse, and sexual abuse were associated with greater suicide risk via greater PTSD symptoms. These findings can be used to better understand responses to CM and IPV and identify pathways leading to suicide, which is essential for developing targeted interventions that correspond with risk profiles.

PMID:41871370 | DOI:10.1177/00332941261436723

Gigascience. 2026 Mar 23:giag030. doi: 10.1093/gigascience/giag030. Online ahead of print.

ABSTRACT

Strain-level metagenomic classification is essential for understanding microbial diversity and functional potential, yet remains challenging, particularly when sample composition is unknown and reference databases are large and redundant. Here we present MADRe, a modular and scalable pipeline for long-read strain-level metagenomic classification based on Metagenome Assembly-Driven Database Reduction. Beyond system-level integration, MADRe introduces statistical strategies that leverage assembly-derived genomic context to guide database reduction and probabilistic read reassignment. Specifically, it combines long-read metagenome assembly, contig-to-reference reassignment using an expectation-maximization framework for reference reduction, and probabilistic read mapping reassignment on a reduced database to achieve sensitive and precise strain-level classification. We extensively evaluated MADRe on simulated datasets, mock communities, and a real anaerobic digester sludge metagenome. Across diverse similarity and coverage conditions, MADRe consistently improves precision by reducing false-positive strain detections. MADRe’s design allows users to apply either the database reduction or read classification step individually. Using only the read classification step shows results on par with other tested tools. MADRe is open source and publicly available at https://github.com/lbcb-sci/MADRe.

PMID:41871361 | DOI:10.1093/gigascience/giag030

Ann Intern Med. 2026 Mar 24. doi: 10.7326/ANNALS-25-02600. Online ahead of print.

ABSTRACT

BACKGROUND: Pegfilgrastim-induced bone pain (PIBP) is common and lacks effective treatment.

OBJECTIVE: To determine whether there is an association between the timing of pegfilgrastim administration and PIBP.

DESIGN: Three-arm randomized controlled trial. (ClinicalTrials.gov: NCT05841186).

SETTING: A tertiary A-level hospital.

PATIENTS: Patients with I to III stage breast cancer who were naive to chemotherapy.

INTERVENTION: Patients were randomly allocated in a 1:1:1 ratio to the 24-hour, 48-hour, or 72-hour group based on the timing of pegfilgrastim administration postchemotherapy.

MEASUREMENTS: The primary end point was the area under the curve (AUC) of the daily worst bone pain score (assessed using the “worst pain” question from the Brief Pain Inventory, a 0 to 10 numerical rating scale [NRS]) for 5 consecutive days in the first chemotherapy cycle. Secondary end points included the incidence of severe bone pain (>5 on the NRS), neutropenia, and febrile neutropenia (FN).

RESULTS: The intention-to-treat analyses included 159 patients, with 53 in each group. For the first cycle, in the 72-hour group, the mean AUC exhibited a statistically significant reduction from 12.74 in the 24-hour group and 14.20 in the 48-hour group to 6.05 (all P < 0.001). Furthermore, the incidence of severe bone pain also declined significantly from 58.5% in the 24-hour group and 66.0% in the 48-hour group to 22.6% in the 72-hour group (all P < 0.001). There was no substantial difference in the incidence of neutropenia among groups, and no patients developed FN.

LIMITATION: Open label, single center, and relatively small sample size.

CONCLUSION: Administration of pegfilgrastim 72 hours postchemotherapy reduced PIBP compared with 24- and 48-hour administration and did not seem to be associated with higher rates of neutropenia or FN.

PRIMARY FUNDING SOURCE: National Natural Science Foundation of China.

PMID:41871353 | DOI:10.7326/ANNALS-25-02600

JMIR Form Res. 2026 Mar 23;10:e85452. doi: 10.2196/85452.

ABSTRACT

BACKGROUND: Clinical decision support systems (CDSSs) are widely used in various health care settings. In Japan, pressure ulcers are becoming a major concern in an aging society due to their increasing prevalence. However, management is often handled by nonspecialists in wound care due to regional disparities in specialist availability.

OBJECTIVE: To provide support for nonspecialists in wound care, we developed a prototype smartphone-based CDSS for pressure ulcer management. The system prompts users to answer questions about the wound’s condition and recommends appropriate ointments and wound dressings by using a safety-first approach. This study aims to evaluate the utility of this system.

METHODS: We conducted a randomized crossover pilot study involving 28 general internal medicine (GIM) physicians. Participants were randomly assigned to group A (intervention-control) or group B (control-intervention). Participants evaluated 10 standardized pressure ulcer photographs and selected the most appropriate ointment and wound dressing for each. The unit of analysis was the individual response to each question (N=280 total observations). We used generalized estimating equations with an exchangeable correlation structure to account for within-subject clustering and adjust for potential period and sequence effects.

RESULTS: The overall correct response rate during the intervention phase was significantly higher than that during the control phase (49.3% vs 4.3%, respectively). After adjusting for clustering and crossover biases, the use of CDSS was associated with a 29.1-fold increase in the odds of a correct response (95% CI 8.2-103; P<.001). Secondary analyses revealed significant improvements in ointment selection (adjusted odds ratio [aOR] 2.4, 95% CI 1.5-3.8; P<.001) and wound dressing selection (aOR 8.9, 95% CI 4.9-16.1; P<.001). However, no significant period (P=.11) or sequence (P=.25) effects were observed for the primary outcome.

CONCLUSIONS: The prototype CDSS improved the accuracy of treatment decisions made by GIM physicians in a pilot study that used photographs and fixed options. Within the parameters of this investigation, CDSS effectively guided participants toward standardized, safety-oriented choices as defined by our scoring criteria.

TRIAL REGISTRATION: UMIN Clinical Trials Registry UMIN000057294; https://tinyurl.com/36a6vvah.

PMID:41871340 | DOI:10.2196/85452

J Med Internet Res. 2026 Mar 23;28:e85583. doi: 10.2196/85583.

ABSTRACT

BACKGROUND: Mobile health (mHealth) apps can address health inequities and enhance access to care for individuals with immunocompromising conditions. Although hundreds of oncology apps exist, research on provider perspectives regarding their use in clinical care remains limited.

OBJECTIVE: This study aimed to describe oncology providers’ recommended apps, mHealth attitudes and beliefs, and perceived barriers to and facilitators of mHealth adoption. Exploratory aims examined differences based on provider type (medical vs psychosocial), provider age (<45 vs ≥45 years old), and patient population (pediatric vs adult).

METHODS: We conducted a cross-sectional survey administered via REDCap (Research Electronic Data Capture) to oncology providers across the United States between June and November 2024. Data were summarized using descriptive statistics. Pearson’s chi-square analyses examined exploratory group differences based on provider type, provider age, and patient age.

RESULTS: Of 188 respondents, the majority self-identified as female (150/188, 79.8%), White (161/188, 85.6%), and non-Hispanic/Latino (174/188, 92.6%). Nearly all providers (178/188, 94.7%) reported either recommending or using mHealth apps with their patients, with primary use for patient-provider communication (139/188, 73.9%). Providers perceived potential benefit across a broad spectrum of holistic care functions. Providers, on average, reported a growth mindset and confidence in their ability to learn mHealth tools and in its potential to improve care access. Key facilitators included alignment with patient needs, increased accessibility, and cost-effectiveness, while barriers included disparities in technology access, digital health literacy, and data security and privacy. Exploratory analyses showed some significant group differences by provider role, provider age, and patient age. Psychosocial providers were significantly more likely to recommend or use apps for pain management (χ²₁=14.34, P<.001, φ=0.28), mental health (χ²₁=50.54, P<.001, φ=0.53), and sleep health (χ²₁=25.47, P<.001, φ= 0.38). Psychosocial providers also perceived higher benefit for sleep health apps (χ²₁=6.40, P=.01, φ=0.19). Medical providers were significantly more likely to perceive medication management apps as potentially beneficial (χ²₁=10.93, P<.001, φ=0.25). Older providers (16/88, 18.2%) and adult care providers (8/32, 25%) were significantly more likely to recommend or use disease management apps compared to younger providers (5/100, 5%; χ²₁=8.20, P=.004, φ=0.21) and pediatric care providers (6/101, 5.9%; χ²₂=9.22, P=.01, Cramer V=0.22), respectively. Pediatric care providers (83/101, 82.2%) were more likely to recommend or use medical team communication apps compared to adult care providers (15/32, 46.9%; χ²₂=15.66, P<.001, Cramer V=0.29).

CONCLUSIONS: Our study underscores the opportunity to develop inclusive mHealth solutions tailored to the diverse needs of individuals across the cancer care continuum, including those in active treatment and survivorship care. Engaging diverse medical and psychosocial providers is essential to inform clinical integration of mHealth technologies in oncology care.

PMID:41871339 | DOI:10.2196/85583