Category: Statistics

Epilepsia. 2024 Nov;65(11):3324-3334. doi: 10.1111/epi.18118. Epub 2024 Sep 18.

ABSTRACT

OBJECTIVE: Phenotypic heterogeneity presents challenges in providing clinical care to patients with pathogenic SCN8A variants, which underly a wide disease spectrum ranging from neurodevelopmental delays without seizures to a continuum of mild to severe developmental and epileptic encephalopathies (DEEs). An important unanswered question is whether there are clinically important subgroups within this wide spectrum. Using both supervised and unsupervised machine learning (ML) approaches, we previously found statistical support for two and three subgroups associated with loss- and gain- of- function vari-ants, respectively. Here, we test the hypothesis that the unsupervised subgroups (U1-U3) are distinguished by differential contributions of developmental and epileptic components.

METHODS: We predicted that patients in the U1 and U2 subgroups would differ in timing of developmental delay and seizure onset, with earlier and concurrent onset of both features for the U3 subgroup. Standard statistical procedures were used to test these predictions, as well as to investigate clinically relevant associations among all five subgroups.

RESULTS: Two-population proportion and Kruskal-Wallis tests supported the hypothesis of a reversed order of developmental delay and seizure onset for patients in U1 and U2, and nearly synchronous developmental delay/seizure onset for the U3 (termed DEE) subgroup. Association testing identified subgroup variation in treatment response, frequency of initial seizure type, and comorbidities, as well as different median ages of developmental delay onset for all five subgroups.

SIGNIFICANCE: Unsupervised ML approaches discern differential developmental and epileptic components among patients with SCN8A-related epilepsy. Patients in U1 (termed developmental encephalopathy) typically gain seizure control yet rarely experience improvements in development, whereas those in U2 (termed epileptic encephalopathy) have fewer if any developmental impairments despite difficulty in achieving seizure control. This understanding improves prognosis and clinical management and provides a framework to discover mechanisms underlying variability in clinical outcome of patients with SCN8A-related disorders.

PMID:39556335 | DOI:10.1111/epi.18118

Egypt Heart J. 2024 Nov 18;76(1):151. doi: 10.1186/s43044-024-00578-z.

ABSTRACT

BACKGROUND: Myocardial deformation by speckle tracking echocardiography provides valuable information on the left ventricular function. The study aims to assess myocardial deformation in terms of left ventricular strain as an indicator of myocardial function in children after cardiac surgery at outpatient follow-up visits.

METHODS: The study design was a prospective observational cross-sectional study that included pediatric patients after biventricular cardiac surgery during the postoperative follow-up visits in the outpatient department. In addition to conventional echocardiographic examination, two-dimensional speckle tracking echocardiography was done to evaluate myocardial deformation in terms of left ventricular strain. Echocardiographic measurements were done offline and were compared to published reference normal values for age. Study subjects were divided according to age at follow-up into four groups (1 month-1 year, 1-2 years, 2-5 years, and 5-11 years).

RESULTS: Over ten months, 100 patients (64 males and 36 females) were included in the study. The median age was 30.8 months (IQR 12.8-65.3 months), the median weight was 11.7 kg (IQR 8-17 kg) and the median duration after surgery was 7.3 months (IQR 3.2-30.8 months). Longitudinal strain values were significantly (p < 0.001) lower than reference values for different age groups. Global circumferential strain showed no significant difference from the reference values. The duration after surgery had a statistically significant effect on longitudinal strain values, with improvement of the strain values with increasing intervals after surgery.

CONCLUSION: Using myocardial deformation method to evaluate cardiac function may detect underlying cardiac function abnormalities even with normal traditional functional parameters, which could have implications for patient management and follow-up.

PMID:39556306 | DOI:10.1186/s43044-024-00578-z

Adv Ther. 2024 Nov 18. doi: 10.1007/s12325-024-03053-y. Online ahead of print.

ABSTRACT

INTRODUCTION: Although clinical, functional, and biomarker data predict asthma exacerbations, newer approaches providing high accuracy of prognosis are needed for real-world decision-making in asthma. Machine learning (ML) leverages mathematical and statistical methods to detect patterns for future disease events across large datasets from electronic health records (EHR). This study conducted training and fine-tuning of ML algorithms for the real-world prediction of asthma exacerbations in patients with physician-diagnosed asthma.

METHODS: Adults with ≥ 2 ICD9/10 asthma codes within 1 year and at least 30 days apart were identified from the Optum Panther EHR database between 2016 and 2023. An emergency department (ED), urgent care, or inpatient visit for asthma, while on systemic administration of corticosteroids, was considered an exacerbation. To predict factors associated with exacerbations in a 6-month study period, clinical information from patients was retrieved in the preceding 6-month baseline period. Clinical information included demographics, lab results, diagnoses, medications, immunizations, and allergies. Three models built using Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM), and Transformers algorithms were trained and tested on independent datasets. Predictions were explained using the SHAP (SHapley Additive exPlanations) library.

RESULTS: Of 1,331,934 patients with asthma, 16,279 (1.2%) experienced ≥ 1 exacerbation. XGBoost was the best predictive algorithm (area under the curve [AUC] = 0.964). Factors associated with exacerbations included a prior history of exacerbation, prednisone usage, high-dose albuterol usage, and elevated troponin I. Reduced probability of exacerbations was associated with receiving inhaled albuterol, vitamins, aspirin, statins, furosemide, and influenza vaccination.

CONCLUSION: This ML-based study on asthma in the real world confirmed previously known features associated with increased exacerbation risk for asthma, while uncovering not entirely understood features associated with reduced risk of asthma exacerbations. These findings are hypothesis-generating and should contribute to ongoing discussion of the strengths and limitations of ML and other supervised learning models in patient risk stratification.

PMID:39556295 | DOI:10.1007/s12325-024-03053-y

Health Econ Rev. 2024 Nov 18;14(1):94. doi: 10.1186/s13561-024-00573-9.

ABSTRACT

OBJECTIVES: Value-based insurance design (VBID) aims to direct consumers’ preferences by incentivizing the use of high-value care and discouraging the use of low-value care. However, consumers often have limited knowledge of health insurance and the health insurance system, possibly distorting their preferences. In this study, we aim to investigate the impact of specific information treatments on consumers’ preferences for VBID.

METHODS: We implemented an information experiment as part of a representative survey on health insurance literacy and preferences for VBID within Switzerland’s choice-based health insurance system. Preferences for VBID were measured through a discrete choice experiment. Cross-sectional data on 6,033 respondents aged 26-75 were analyzed using descriptive statistics and mixed logit regressions.

RESULTS: Respondents showed strong preferences for their current health insurance instead of VBID alternatives. A general description of current regulations on cost-sharing, drug disbursement, and monthly premiums significantly increased preferences for VBID (p < 0.01). Pointing respondents specifically to VBID further reduced the opposition against VBID plans. At the same time, there is evidence for anchoring effects in copayments after receiving the information treatments, irrespective of the value of the care.

CONCLUSIONS: The results of this study highlight that individuals are susceptible to provided information about health insurance when building their preferences for VBID. One potential explanation is limited health insurance literacy, implying that tailored communication strategies may be needed to improve insurance decision-making.

JEL CLASSIFICATION: I11, I13.

PMID:39556285 | DOI:10.1186/s13561-024-00573-9

Hernia. 2024 Nov 18;29(1):22. doi: 10.1007/s10029-024-03219-1.

ABSTRACT

PURPOSE: To evaluate the effectiveness of prophylactic mesh placement in reducing the incidence of parastomal hernias following colostomy, ileostomy, and ileal conduit formation.

METHODS: A systematic review identified relevant studies evaluating parastomal hernia incidence with prophylactic mesh use during stoma formation. Pairwise meta-analysis and network meta-analysis using Bayesian modeling were performed.

RESULTS: 25 studies, consisting of 16 randomized control trials (RCT), 6 follow up studies, and 3 retrospective cohort studies, were included. Prophylactic mesh led to significantly fewer parastomal hernias beyond 6 months follow-up (OR 0.43, 95% CI 0.33-0.58). Hernias were reduced with mesh for both ileal conduits and colostomies. When analyzing hazard ratios (HRs), only 6 studies were included, and a statistically significant difference was observed among both randomized controlled trials (RCTs) (HR 0.75 [0.53, 0.92], p = 0.01) and non-RCTs (HR 0.57 [0.36, 0.92], p = 0.02). Network meta-analysis found the retromuscular approach with mesh had the lowest hernia rate. Regression was non-significant for variations between study types.

CONCLUSION: This meta-analysis demonstrated prophylactic mesh placement during ostomy creation significantly reduced parastomal hernia risk, more prominently beyond 6 months, consistently across randomized trials and observational studies for urologic and gastrointestinal ostomies. The retromuscular technique was most effective.

PMID:39556272 | DOI:10.1007/s10029-024-03219-1

Clin Child Fam Psychol Rev. 2024 Nov 18. doi: 10.1007/s10567-024-00497-0. Online ahead of print.

ABSTRACT

As attachment-based interventions can improve caregiver-infant relationships and their subsequent psychological outcomes, the identification of relevant and effective interventions can facilitate their implementation into clinical practice. This systematic review aimed to a) provide an overview of manualised attachment-based interventions, without video-feedback as the main component, for caregivers and infants from conception to two years, and b) determine which of these interventions were effective in demonstrating improvements in caregiver-infant relational outcomes. To identify eligible interventions and their empirical evidence base, two search stages were conducted for 1) relevant interventions and 2) studies of interventions identified in the first stage that focussed on caregiver-infant relational outcomes. All studies included in Stage 2 were quality assessed and findings analysed. Twenty-six interventions were eligible for inclusion at Stage 1 but studies reporting on relational outcomes were identified for 16 interventions only. Forty studies reporting on those 16 interventions met inclusion criteria and were synthesised at Stage 2. Most studies were of good quality. Observer-rated measures were used in 90% of studies. There was evidence for these interventions in relation to improving caregiver-infant relational outcomes: 80% of studies reported a statistically significant positive change in a relational outcome for the intervention compared to pre-intervention or control group. The most promising evidence was identified for Attachment and Biobehavioral Catch-Up (ABC), Minding the Baby (MTB) and Circle of Security (COS). This systematic review offers guidance to healthcare professionals, commissioners and policymakers within perinatal sectors in relation to the training, delivery and implementation of evidenced manualised attachment-based interventions.

PMID:39556257 | DOI:10.1007/s10567-024-00497-0

J R Soc Med. 2024 Nov 18:1410768241297833. doi: 10.1177/01410768241297833. Online ahead of print.

ABSTRACT

OBJECTIVES: Using electronic health records, we derived and internally validated a prediction model to estimate risk factors for long COVID and predict individual risk of developing long COVID.

DESIGN: Population-based, retrospective cohort study.

SETTING: Scotland.

PARTICIPANTS: Adults (≥18 years) with a positive COVID-19 test, registered with a general medical practice between 1 March 2020 and 20 October 2022.

MAIN OUTCOME MEASURES: Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) for predictors of long COVID, and patients’ predicted probabilities of developing long COVID.

RESULTS: A total of 68,486 (5.6%) patients were identified as having long COVID. Predictors of long COVID were increasing age (aOR: 3.84; 95% CI: 3.66-4.03 and aOR: 3.66; 95% CI: 3.27-4.09 in first and second splines), increasing body mass index (BMI) (aOR: 3.17; 95% CI: 2.78-3.61 and aOR: 3.09; 95% CI: 2.13-4.49 in first and second splines), severe COVID-19 (aOR: 1.78; 95% CI: 1.72-1.84); female sex (aOR: 1.56; 95% CI: 1.53-1.60), deprivation (most versus least deprived quintile, aOR: 1.40; 95% CI: 1.36-1.44), several existing health conditions. Predictors associated with reduced long COVID risk were testing positive while Delta or Omicron variants were dominant, relative to when the Wild-type variant was dominant (aOR: 0.85; 95% CI: 0.81-0.88 and aOR: 0.64; 95% CI: 0.61-0.67, respectively) having received one or two doses of COVID-19 vaccination, relative to unvaccinated (aOR: 0.90; 95% CI: 0.86-0.95 and aOR: 0.96; 95% CI: 0.93-1.00).

CONCLUSIONS: Older age, higher BMI, severe COVID-19 infection, female sex, deprivation and comorbidities were predictors of long COVID. Vaccination against COVID-19 and testing positive while Delta or Omicron variants were dominant predicted reduced risk.

PMID:39556251 | DOI:10.1177/01410768241297833

Acta Diabetol. 2024 Nov 18. doi: 10.1007/s00592-024-02415-w. Online ahead of print.

ABSTRACT

AIMS: Diabetes patients are at a higher risk of fractures, and glucagon-like peptide-1 receptor agonists (GLP-1RAs) have been suggested to positively impact on bone metabolism. We aim to provide a comprehensive assessment of fracture events associated with GLP-1RAs based on pharmacovigilance data.

METHODS: In this study, fracture-related adverse events (AEs) associated with GLP-1RAs and other commonly used glucose-lowering drugs were identified from Food and Drug Administration Adverse Event Reporting System (FAERS) database (2004-2022). The reporting odds ratio (ROR) and adjusted ROR (adj. ROR) were used to compare the reporting of fracture-related AEs associated with insulin, GLP-1RAs, and Non GLP-1RAs, in patients with diabetes through two scenarios. This involved separately comparing each glucose-lowering drug to all other medications used in diabetic patients and reiterating after excluding insulin cases.

RESULTS: A total of 490,107 AE reports for patients with diabetes were identified and 98, 625 of them were for GLP-1RAs. Among all diabetes drugs, GLP-1RAs had the lowest reporting of any fracture-related AEs [adj. ROR = 0.44 (0.40-0.47)], consistent across osteoporotic fracture [adj. ROR = 0.39 (0.34-0.45)] and hip fracture [adj. ROR = 0.34 (0.28-0.41)]. Among GLP-1RA agents, albiglutide was associated with the lowest adj. ROR [0.11 (0.05-0.21)] for any fracture-related AEs. After excluded all insulin reports, GLP-1RAs retained a significantly lower adj. ROR towards any fracture [adj. ROR = 0.45 (0.40-0.50)], osteoporotic fracture [adj. ROR = 0.44 (0.37-0.52)], and hip fracture [adj. ROR = 0.43 (0.33-0.54)].

CONCLUSION: In a real-world pharmacovigilance setting, GLP-1RAs were associated with lower reporting of fracture-related AEs, indicating the protective effect of GLP-1RAs against fractures.

PMID:39556224 | DOI:10.1007/s00592-024-02415-w

Acta Diabetol. 2024 Nov 18. doi: 10.1007/s00592-024-02412-z. Online ahead of print.

ABSTRACT

BACKGROUND: The triglyceride glucose (TyG) index serves as a dependable surrogate biomarker for evaluating insulin resistance. However, the role of the TyG index in patients with diabetes mellitus who also suffer from acute renal failure warrants further investigation. This study sought to investigate the association between the TyG index and the incidence of acute renal failure in individuals with diabetes.

METHODS: This study utilized data from the MIMIC-IV database, categorizing patients into tertiles according to their TyG index. Employing multivariate logistic regression models, we analyzed the relationship between the TyG index and the occurrence of acute renal failure among diabetic patients. To assess non-linear relationships, restricted cubic splines were utilized, and upon detection of non-linearity, a recursive algorithm was implemented to determine inflection points.

RESULTS: The study comprised a total of 1074 participants diagnosed with diabetes mellitus. In the model adjusted for all covariates, the odds ratio (OR) for the association between the TyG index and acute renal failure, accompanied by a 95% confidence interval (CI), was 1.22 (0.82, 1.82), which did not reach statistical significance. However, analysis using restricted cubic splines revealed a U-shaped relationship between the TyG index and acute renal failure, with an inflection point at 9.26. The relationship between the TyG index and acute renal failure was inverse before reaching the inflection point and became directly proportional thereafter, with an OR (95% CI) of 1.86 (1.12, 3.09) after the point.

CONCLUSION: In individuals diagnosed with diabetes mellitus, our analysis revealed a non-linear relationship between the TyG index and the incidence of acute renal failure. Beyond the inflection point, elevated TyG index levels were markedly linked to a higher prevalence of acute renal failure.

PMID:39556222 | DOI:10.1007/s00592-024-02412-z

Clin Res Cardiol. 2024 Nov 18. doi: 10.1007/s00392-024-02575-3. Online ahead of print.

ABSTRACT

AIM: Delirium in patients treated in the intensive care unit (ICU) is linked to adverse outcome, according to previous observations. However, data on patients recovering after cardiac arrest are sparse. The aim of this study was to assess incidence, risk factors, and outcome of patients with delirium after cardiac arrest in the Freiburg Delirium Registry (FDR).

METHODS: In this retrospective registry study, all patients after cardiac arrest treated in the Freiburg University Medical Center medical ICU between 08/2016 and 03/2021 were included. Delirium was diagnosed using the Nursing Delirium screening scale (NuDesc), assessed three times daily. Favorable neurological outcome was defined as cerebral performance category (CPC) score at ICU discharge ≤ 2.

RESULTS: Two hundred seventeen patients were included and among them, delirium was detected in one hundred ninety-nine (91.7%) patients. Age was independently associated with the incidence of delirium (p = 0.003), and inversely associated with the number of delirium-free days (p < 0.001). Favorable neurological outcome was present in 145/199 (72.9%) with, and 17/18 (94.4%) patients without delirium (p = 0.048). While the incidence of delirium was not independently associated with a favorable neurologic outcome, the number of delirium-free days strongly predicted the primary endpoint [OR 2.14 (1.73-2.64), p > 0.001].

CONCLUSION: Delirium complicated the ICU course in almost all patients after cardiac arrest. The number of delirium-free days was associated with favorable outcome while incidence of delirium itself was not.

PMID:39556214 | DOI:10.1007/s00392-024-02575-3