Tag: statistics

Age Ageing. 2025 Aug 29;54(10):afaf285. doi: 10.1093/ageing/afaf285.

ABSTRACT

OBJECTIVE: Older adults face high risk of falls. We developed an electronic-health-record (EHR) based machine-learning (ML) model to predict 1-year risk of fall in older adults for pre-emptive intervention.

METHODS: We included 4 902 161 records from 1 142 000 adults aged ≥65 years who attended the Hong Kong Hospital Authority (HA) facilities in 2013-2017. We included 260 predictors including demographics, in-patient/out-patient admissions, emergency department (ED) attendance, complications, medications and laboratory tests during 1-year period to predict fall events based on diagnostic codes in the ensuing 12 months. The cohort was randomly split into training, testing and internal validation sets in a 7:2:1 ratio. We evaluated the performance of six ML-algorithms.

RESULTS: 67 163 fall events were accrued with the XGBoost model having the best performance in the validation set (area-under-the-receiver-operating-characteristic-curve [AUROC] = 0.979, area-under-the-precision-recall-curve [AUPRC] = 0.764; positive-predictive-value [PPV] = 0.614) versus logistic-regression model (AUROC = 0.885, AUPRC = 0.169; PPV = 0.210). The top 30 predictors included number of ED attendance, fasting plasma glucose, number and types of outpatient appointments, ED triage category of ‘urgent’, number of admissions and stay, age, residential districts, history of fall and medication use with an AUROC of 0.939 in a validation cohort of patients with diabetes. In an age- and sex-matched sub-cohort, compared to the widely-used Morse Fall Score, XGBoost model had higher sensitivity (0.569-versus-0.139) with optimal balance of identifying positive cases whilst simultaneously minimising false positives and false negative (F1 score: 0.626-versus-0.555).

CONCLUSIONS: Our ML-model highlights the utility of EHR in identifying high-risk individuals for falls, supporting integrating into the EHR system for targeted preventive actions.

PMID:41066674 | DOI:10.1093/ageing/afaf285

Epilepsia. 2025 Oct 9. doi: 10.1111/epi.18667. Online ahead of print.

ABSTRACT

OBJECTIVE: This study was undertaken to develop a framework for focal cortical dysplasia (FCD) detection using surface-based morphometric (SBM) analysis and machine learning (ML) applied to three-dimensional (3D) magnetic resonance fingerprinting (MRF).

METHODS: We included 114 subjects (44 patients with medically intractable focal epilepsy and FCD, 70 healthy controls [HCs]). All subjects underwent high-resolution 3-T MRF scans generating T1 and T2 maps. All patients had clinical T1-weighted (T1w) images; 35 also had 3D fluid-attenuated inversion recovery (FLAIR). A 3D region of interest (ROI) was manually created for each lesion. All maps/images and lesion ROIs were registered to T1w images. Surface-based features were extracted following the Multi-center Epilepsy Lesion Detection pipeline. Features were normalized using intrasubject, interhemispheric, and intersubject z-scoring. A two-stage ML approach was applied: a vertexwise neural network classifier for lesional versus normal vertices using T1w/MRF/FLAIR features, followed by a clusterwise Random Undersampling Boosting classifier to suppress false positives (FPs) based on cluster size, prediction probabilities, and feature statistics. Leave-one-out cross-validation was performed at both stages.

RESULTS: Using T1w features, sensitivity was 70.4% with 11.6 FP clusters/patient and 4.1 in HCs. Adding MRF reduced FPs to 6.6 clusters/patient and 1.5 in HCs, with 68.2% sensitivity. Combining T1w, MRF, and FLAIR achieved 71.4% sensitivity, with 4.7 FPs/patient and 1.1 in HCs. Detection probabilities were significantly higher for true positive clusters than FPs (p < .001). Type II showed higher detection rates than non-type II. Magnetic resonance imaging (MRI)-positive patients showed higher detection rates and fewer FPs than MRI-negative patients. Seizure-free patients demonstrated higher detection rates than non-seizure-free patients. Subtyping accuracy was 80.8% for non-type II versus type II, and 68.4% for IIa versus IIb, although limited by small sample size. The transmantle sign was present in 61.5% of IIb and 40% of IIa cases.

SIGNIFICANCE: We developed an ML framework for FCD detection integrating SBM with clinical MRI and MRF. Advances include improved FP control and enhanced subtyping; selected model outputs may provide indicators of detection confidence and seizure outcome.

PMID:41066149 | DOI:10.1111/epi.18667

Spine (Phila Pa 1976). 2025 Oct 8. doi: 10.1097/BRS.0000000000005530. Online ahead of print.

ABSTRACT

STUDY DESIGN: Retrospective cohort study with a prospective qualitative component.

OBJECTIVE: To identify themes contributing to satisfaction following lumbar surgery in patients who improved on global rating of change (GRC) but did not achieve ODI MCID or PASS.

SUMMARY OF BACKGROUND DATA: Outcomes after lumbar surgery are often evaluated using Oswestry Disability Index (ODI), minimal clinically important difference (MCID), and patient acceptable symptom state (PASS). However, these quantitative metrics fail to capture “qualitative” measures of satisfaction.

MATERIALS AND METHODS: Patient reported outcomes (PROMs) were retrospectively collected and evaluated at early (<6 mo) and late (≥6 mo) follow-up. Wilcoxon signed-rank and Fisher’s exact tests were used to compare PROMs, MCID, and GRC satisfaction, defined as answering “much better” or “slightly better” on≥6 months GRC. Prospective interviews explored various physical, emotional, and social “themes” using grounded theory. “Thematic saturation” was determined when new patient responses ceased to generate unique themes.

RESULTS: 43 patients reported satisfaction and consented for interview. Statistically significant improvements for VAS back and leg were reported at early and late follow-up, but not for ODI, SF12 PCS, or SF12 MCS. Interviews generated five themes, related to physical independence, healthcare provider and institutional experience, mental wellbeing and pain reduction, personal support, and environmental factors. Thematic saturation occurred after 32 patients. Patients identified improvements in pain, function, and mental wellness, along with positive provider interactions, reputation, social support, and environment, as key contributors to postoperative satisfaction.

CONCLUSION: Satisfaction after lumbar surgery is impacted by various qualitative factors related to their healthcare team, social support, and other environmental factors, as well as improvements in mental and physical function. These findings underscore key elements beyond the ODI that surgeons should consider to optimize postoperative satisfaction. They may also support the development of new clinical tools to supplement the ODI and existing PROMs by capturing qualitative aspects of satisfaction not routinely assessed.

LEVEL OF EVIDENCE: 3.

PMID:41066143 | DOI:10.1097/BRS.0000000000005530

Br J Surg. 2025 Oct 3;112(10):znaf211. doi: 10.1093/bjs/znaf211.

ABSTRACT

BACKGROUND: A body of evidence supports a link between metabolic bariatric surgery (MBS) and alcohol use disorder (AUD), while the possible contribution to alcohol-related mortality remains unclear. The aim of this study was to examine the association between MBS and the risk of AUD and alcohol-related mortality over up to 35 years.

METHODS: The Swedish Obese Subjects (SOS) study enrolled 2007 participants with severe obesity who underwent MBS and 2040 matched controls (median follow-up 25.2 years). Patients in the surgery group underwent gastric bypass (GBP; 266 patients), gastric banding (376 patients), or vertical banded gastroplasty (VBG; 1365 patients). The matched controls received the customary treatment for severe obesity at their primary healthcare centres. Data on AUD diagnoses and alcohol-related mortality were captured from the Swedish National Patient Register and the Swedish Cause of Death Register respectively.

RESULTS: During long-term follow-up, a significant difference in the incidence of AUD was found across surgery groups (log rank P < 0.001). Patients who underwent GBP exhibited the highest AUD risk (adjusted HR (HRadj) 5.07 (95% c.i. 3.11 to 8.25); P < 0.001), followed by patients who underwent VBG (HRadj 2.28 (95% c.i. 1.56 to 3.34); P < 0.001) and patients who underwent gastric banding (HRadj 2.34 (95% c.i. 1.37 to 4.01); P = 0.002), compared with usual obesity care. Alcohol-related mortality was significantly elevated after GBP (adjusted sub-HR (sub-HRadj) 6.18 (95% c.i. 2.48 to 15.40); P < 0.001) and VBG (sub-HRadj 3.56 (95% c.i. 1.79 to 7.08); P < 0.001) compared with usual obesity care. Mortality after gastric banding was also elevated, but did not reach statistical significance (sub-HRadj 2.52 (95% c.i. 0.89 to 7.15); P = 0.082).

CONCLUSION: Effective management of alcohol-related complications in MBS patients requires preoperative risk assessment, postoperative monitoring, and access to targeted interventions for AUD.

PMID:41066138 | DOI:10.1093/bjs/znaf211

JAMA Oncol. 2025 Oct 9. doi: 10.1001/jamaoncol.2025.3863. Online ahead of print.

ABSTRACT

IMPORTANCE: Substantial improvements in childhood cancer survival have created a critical need to address serious long-term health complications, such as valvular heart disease (VHD).

OBJECTIVE: To identify treatment-related risk factors for VHD in a large European cohort of long-term childhood cancer survivors.

DESIGN, SETTING, AND PARTICIPANTS: This nested case-control study used data from the PanCareSurFup (PanCare Childhood and Adolescent Cancer Survivor Care and Follow-Up Studies) and ProCardio cohorts, including detailed radiation dose reconstruction and chemotherapy exposure, for childhood cancer survivors from 7 European countries, diagnosed between 1940 and 2009, who survived at least 5 years after cancer diagnosis. Case patients, defined as having symptomatic VHD, were matched with controls 1:2 by subcohort, sex, age at cancer diagnosis, and calendar year of initial diagnosis. Data were analyzed from October 2023 to June 2025.

EXPOSURES: Doses were calculated by performing a whole-body dosimetric reconstruction using a voxel-based anthropomorphic phantom with more than 200 delineated anatomic structures or substructures. Cumulative dose to cytotoxic agents was also assessed.

MAIN OUTCOME AND MEASURE: Development of symptomatic VHD (grade ≥3 per the Common Terminology and Criteria for Adverse Events, version 4.03).

RESULTS: Of the 225 cases, 136 participants (60.4%) were male, and 195 (86.7%) were diagnosed with VHD beyond 20 years from childhood cancer. Survivors receiving a mean heart radiation therapy (RT) dose of 5 to less than 15 Gy had an increased risk of VHD (odds ratio [OR], 4.7; 95% CI, 2.1-10.7) compared to those without heart RT, with higher risk when more than half of the heart was exposed. The heart RT dose response appeared exponential, with the OR being 104.1 (95% CI, 27.8-389.6) for mean heart dose of 30 Gy or more, increasing considerably with follow-up from 6.0 (95% CI, 1.4-26.5) after 5 to 19 years to 71.4 (95% CI, 20.4-250.0) after 30 or more years. Cumulative anthracycline doses of 400 mg/m2 or higher were also associated with increased VHD risk (OR, 3.8; 95% CI, 1.4-10.3), showing an exponential dose-response pattern. Cumulative exposure to platinum agents was associated with VHD risk in a linear manner. No statistically significant associations were found for other chemotherapy agents or radiation to the spleen.

CONCLUSION AND RELEVANCE: In this case-control study, heart RT, anthracyclines, and platinum agents were associated with increased VHD risk in childhood cancer survivors. Risks from both RT and anthracyclines were amplified with age and follow-up, underscoring the need for long-term cardiac surveillance.

PMID:41066131 | DOI:10.1001/jamaoncol.2025.3863

JAMA Netw Open. 2025 Oct 1;8(10):e2536364. doi: 10.1001/jamanetworkopen.2025.36364.

ABSTRACT

IMPORTANCE: Individuals with intellectual disabilities (IDs) are at increased risk for mental health problems, including depression. However, access to effective therapeutic interventions is often limited.

OBJECTIVE: To evaluate the feasibility, acceptance, and efficacy of a self-help smartphone app designed to reduce depressive symptoms and improve self-esteem and quality of life in individuals with IDs.

DESIGN, SETTING, AND PARTICIPANTS: In this 2-arm randomized clinical trial, adults with IDs and depressive symptoms were enrolled online in Germany between April 1 and August 10, 2023. Of the 135 individuals who accessed the survey, 99 met the eligibility criteria. Participants were recruited via social media, third parties (eg, care institutions), and workplaces. Data were collected at baseline and after the intervention. Statistical analyses included complete case and intention-to-treat (ITT) approaches.

INTERVENTION: Participants were randomly assigned (1:1) to either the intervention group using a smartphone app in easy-to-read German mainly on cognitive behavioral therapy or a waiting list control group. Participants in both groups continued to receive care as usual, which may include routine psychosocial support, daily structure provided by caregivers, and access to general health services.

MAIN OUTCOMES AND MEASURES: The primary outcome was a reduction in depressive symptoms, measured by the Glasgow Depression Scale for People With a Learning Disability. Secondary outcomes were improvements in self-esteem, quality of life, and participant satisfaction.

RESULTS: Among the 99 participants (mean [SD] age, 34.9 [12.6] years; 54 [54.5%] female), 92 completed the postintervention assessment. The intervention group showed a significant reduction in depressive symptoms compared with the control group in ITT analyses (F1,97 = 7.52; P = .007; ηp2 = 0.072; medium effect size), as well as significant improvements in quality of life (F1,97 = 5.09; P = .03; ηp2 = 0.050; small to medium effect size) and in self-esteem (F1,97 = 17.94; P < .001; ηp2 = 0.156; large effect size). Complete case analyses yielded consistent results on most outcome measures. High satisfaction ratings were reported.

CONCLUSIONS AND RELEVANCE: In this randomized clinical trial, a self-guided smartphone app demonstrated efficacy in reducing depressive symptoms and enhancing quality of life and self-esteem among individuals with IDs. The findings suggest that smartphone-based interventions can provide effective support for this underserved population.

TRIAL REGISTRATION: German Clinical Trials Register Identifier: DRKS00030858.

PMID:41066124 | DOI:10.1001/jamanetworkopen.2025.36364

JAMA Netw Open. 2025 Oct 1;8(10):e2536655. doi: 10.1001/jamanetworkopen.2025.36655.

ABSTRACT

IMPORTANCE: Responders involved in rescue and recovery operations after the collapse of the World Trade Center (WTC) on September 11, 2001, were exposed to airborne carcinogens.

OBJECTIVES: To examine the incidence of lung cancer after the WTC attacks and to compare the incidence of lung cancer among responders with varying degrees of exposure severity.

DESIGN, SETTING, AND PARTICIPANTS: In this prospective cohort study, data were collected between July 1, 2012, and December 31, 2023, from individuals who were enrolled in a medical monitoring program available to WTC responders residing on Long Island, New York. This study was restricted to people who survived and were followed up for incident lung cancer after a 10-year latency period.

EXPOSURES: Types and durations of exposures were based on responses to a detailed questionnaire about on-site work conditions, which included information about the type and duration of work, smells, and sights while working; exposure to dust; and the use of protective equipment. World Trade Center exposure characteristics and overall severity were measured as mild, moderate, and severe exposure using a validated approach.

MAIN OUTCOMES AND MEASURES: The incidence of lung cancer was the primary outcome. Diagnosis of lung cancer was ascertained following a standardized approach by trained clinicians, and diagnoses were verified by clinicians at the Centers for Disease Control and Prevention. Cox proportional hazards regression was used to estimate multivariable-adjusted hazard ratios.

RESULT: Among 12 334 eligible responders (mean [SD] age at study inclusion, 49.3 [10.2] years; 11 213 men [90.9%]), 118 incident lung cancers were identified between July 1, 2012, and December 31, 2023 (incidence rate, 8.7/10 000 person-years [95% CI, 7.3-10.5 person-years]). When compared with mild exposures, the incidence of lung cancer was higher among moderately (adjusted hazard ratio [AHR], 1.86 [95% CI, 1.19-2.91]; P = .007) and severely (AHR, 2.90 [95% CI, 1.69-4.99]; P < .001) exposed groups. Specific WTC exposures, including smelling fumes (AHR, 1.05 [95% CI, 1.01-1.09]; P = .007) or sewage (AHR, 1.03 [95% CI, 1.01-1.05]; P = .004), were also associated with higher incidence of lung cancer after adjusting for demographics and measures of tobacco use.

CONCLUSIONS AND RELEVANCE: In this cohort study of WTC responders, the incidence of lung cancer was higher among those with greater exposure severity. Future studies may investigate specific WTC exposures and histologic changes and clarify the role of WTC exposure for prognosis.

PMID:41066122 | DOI:10.1001/jamanetworkopen.2025.36655

JAMA Netw Open. 2025 Oct 1;8(10):e2536771. doi: 10.1001/jamanetworkopen.2025.36771.

ABSTRACT

IMPORTANCE: Narcolepsy is a sleep disorder potentially affecting mortality, yet evidence on this association remains sparse.

OBJECTIVE: To examine whether narcolepsy is associated with an increased risk of all-cause and cause-specific mortality.

DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study used data from the Taiwan National Health Insurance Research Database (NHIRD) from 2001 to 2021, with patients followed up until death or December 31, 2022. Patients were aged 6 years or older with 2 or more narcolepsy diagnoses from psychiatrists or neurologists. Controls were selected from the NHIRD as a population-based sample. Controls without narcolepsy were matched in a 1:4 ratio on sex and birth date (±6 months). Sibling controls were siblings without narcolepsy. Statistical analysis was performed from January to April 2025.

EXPOSURES: Clinical narcolepsy diagnosis, confirmed via NHIRD records (International Classification of Diseases, Ninth Revision, Clinical Modification code 347 or International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Clinical Modification code G47.4).

MAIN OUTCOMES AND MEASURES: The primary outcome was all-cause mortality, measured as hazard ratios (HRs) using Cox proportional hazards regression, adjusted for birth year, sex, income, urbanization, and Charlson Comorbidity Index. Secondary outcomes included cause-specific mortality (natural, unnatural, accidents, suicides).

RESULTS: Of 3187 patients with narcolepsy (mean [SD] age, 29.5 [16.1] years; 1674 male patients [52.5%]) and 12 748 controls (mean [SD] age, 29.5 [16.1] years; 6696 male patients [52.5%]), 132 patients with narcolepsy and 456 controls died. Psychiatric comorbidities, especially depression (1167 of 3187 [36.6%] vs 861 of 12 748 [6.8%]) and anxiety (1054 of 3187 [33.1%] vs 853 of 12 748 [6.7%]), were more common in the narcolepsy group than in the control group. Crude all-cause mortality rates were 44.3 per 10 000 person-years among patients with narcolepsy and 38.1 per 10 000 person-years among controls. All-cause mortality was not increased among patients with narcolepsy (HR, 0.96; 95% CI, 0.79-1.17). There was no increase among patients with narcolepsy in cause-specific mortality for natural causes (HR, 0.90; 95% CI, 0.73-1.11), unnatural causes, (HR, 1.41; 95% CI, 0.83-2.40), accidents (HR, 1.37; 95% CI, 0.64-2.95), and suicides (HR, 1.41; 95% CI, 0.62-3.22). The sibling cohort analysis similarly demonstrated no significantly increased risk among patients with narcolepsy of all-cause mortality (HR, 1.14; 95% CI, 0.63-2.06) or cause-specific mortality from natural causes (HR, 0.66; 95% CI, 0.28-1.56), unnatural causes (HR, 2.08; 95% CI, 0.87-4.98), accidents (HR, 1.61; 95% CI, 0.48-5.37), or suicides (HR, 3.43; 95% CI, 0.88-13.28).

CONCLUSIONS AND RELEVANCE: In this cohort study of Taiwanese residents, narcolepsy was not associated with excess all-cause or cause-specific mortality. These findings reduce immediate clinical concern, but the wide 95% CIs suggest that a modest increase in risk cannot be excluded; replication in other populations with longer follow-up is warranted.

PMID:41066120 | DOI:10.1001/jamanetworkopen.2025.36771

Surg Laparosc Endosc Percutan Tech. 2025 Oct 9. doi: 10.1097/SLE.0000000000001410. Online ahead of print.

ABSTRACT

BACKGROUND: We designed a study to determine the impact of transversalis fascia repair (TFR) during TEP surgery for inguinal hernias on hospital stay duration, recurrence rates, quality of life, and related adverse outcomes in a randomized, double-blinded, controlled clinical trial. We screened patients presenting with inguinal hernias requiring elective surgery.

METHODS: Eligible patients were randomly allocated into 2 groups: the TFR group, whose inguinal ring defect was narrowed; and the Routine Treatment Group (RTG) group, whose defect left unchanged. The prespecified primary outcomes were the length of hospital stay and the time of surgery. Patients were followed for 6 months to be evaluated regarding the Visual Analogue Scale (VAS), the rate of recurrence and seroma formation, and their Carolina Comfort Scale (CCS) hernia-specific quality of life.

RESULTS: A total of 60 patients were allocated to study arms with no significant differences in the baseline characteristics. The intervention did not have a significant impact on the duration of hospitalization. However, the operation time in the TFR group was significantly longer than in the RTG group (Cohen’s d=-1.13, 95% CI: -1.67 to -0.58, P<0.001). In contrast, no statistically or clinically significant differences were noted between the groups concerning postoperative pain, analgesic usage, or rates of acute and chronic pain. Furthermore, the TFR group had a much lower risk of seroma formation during the first week after surgery compared with the routine nonclosing defect group, showing an almost 80% lower risk of seroma formation. However, this finding did not reach statistical significance.

CONCLUSION: Although the defect-closing approach resulted in longer operation times, our study did not demonstrate any beneficial effects on hospital stay duration, postoperative pain, or quality of life. However, this approach may reduce seroma formation in the first week postsurgery, which should be confirmed in future meta-analyses.

TRIAL REGISTRATION: This trial was prospectively registered on the Iranian Registry of Clinical Trials on February 29, 2024 (IRCT20180312039067N2).

PMID:41066116 | DOI:10.1097/SLE.0000000000001410

JAMA Oncol. 2025 Oct 9. doi: 10.1001/jamaoncol.2025.3875. Online ahead of print.

ABSTRACT

IMPORTANCE: Inherited pathogenic variants (PVs) in known predisposition genes can greatly increase breast cancer risk, but the combined impact of PV status, family history, and other factors on breast cancer risk in the general US population has not been well described.

OBJECTIVE: To evaluate population-based breast cancer risk estimates for those with established PVs overall and stratified by first-degree family history of breast cancer and other factors.

DESIGN, SETTING, AND PARTICIPANTS: This study used pooled data from 13 US-based breast cancer case-control studies participating in the Cancer Risk Estimates Related to Susceptibility (CARRIERS) consortium. Enrollment for individual studies occurred between 1976 and 2013, and results are based on data released March 2023, with analyses conducted from June 2022 to July 2025.

EXPOSURES: PVs, breast cancer family history, self-reported race and ethnicity, and established risk factors.

MAIN OUTCOMES AND MEASURES: Breast cancer rate ratios for PVs in 7 genes were estimated from the CARRIERS consortium. PV status and incidence and mortality statistics were combined using the Individualized Coherent Absolute Risk Estimation (iCARE) model to estimate conditional cumulative breast cancer risks and 95% CIs, stratified by family history and standardized to the US population. Models that incorporated population-based data and published estimates for established epidemiologic risk factors were also evaluated.

RESULTS: A total of 67 692 women were studied, including 33 841 who were diagnosed with breast cancer. PVs in ATM, BRCA1, BRCA2, CHEK2, and PALB2 were strongly associated with breast cancer risk, with BRCA1 and PALB2 PVs showing evidence of heterogeneity by family history. In models considering PVs, family history, and established risk factors, the estimated cumulative risks of breast cancer by age 50 years ranged from 2.4% (95% CI, 2.4-2.4) in women with no PVs and no family history to 35.5% (95% CI, 21.6-55.1) in PALB2 PV carriers with a family history. Among women who have not been diagnosed with breast cancer by age 50 years, the cumulative risk of breast cancer by age 80 years ranged from 11.1% (95% CI, 11.0-11.2) in noncarriers with no family history to 70.5% (95% CI, 52.8-83.5) for PALB2 carriers with a family history. PV-specific cumulative risk estimates varied across subgroups defined by race and ethnicity and potentially modifiable epidemiologic risk factors.

CONCLUSIONS AND RELEVANCE: In this study, population-based estimates of cumulative breast cancer risk for established PVs, as informed by the CARRIERS case-control sample, varied by family history and potentially modifiable risk factors. These estimates provide guidance for identifying individuals who will most benefit from enhanced screening and prevention strategies.

PMID:41066089 | DOI:10.1001/jamaoncol.2025.3875