Optimizing and Predicting Antidepressant Efficacy in Patients with Major Depressive Disorder Using Multi-Omics Analysis and the Opade AI Prediction Tools.

According to the World Health Organization (WHO), major depressive disorder (MDD) is the fourth leading cause of disability worldwide and the second most common disease after cardiovascular events. Approximately 280 million people live with MDD, with incidence varying by age and gender (female to male ratio of approximately 2:1). Although a variety of antidepressants are available for the different forms of MDD, there is still a high degree of individual variability in response and tolerability. Given the complexity and clinical heterogeneity of these disorders, a shift from "canonical treatment" to personalized medicine with improved patient stratification is needed. OPADE is a non-profit study that researches biomarkers in MDD to tailor personalized drug treatments, integrating genetics, epigenetics, microbiome, immune response, and clinical data for analysis. A total of 350 patients between 14 and 50 years will be recruited in 6 Countries (Italy, Colombia, Spain, The Netherlands, Turkey) for 24 months. Real-time electroencephalogram (EEG) and patient cognitive assessment will be correlated with biological sample analysis. A patient empowerment tool will be deployed to ensure patient commitment and to translate patient stories into data. The resulting data will be used to train the artificial intelligence/machine learning (AI/ML) predictive tool.

Out-of-pocket expenses and parent reported quality of life in children with cow's milk protein allergy in Bogotá, Colombia.

OBJECTIVE: To identify the out-of-pocket expenses and parent-reported quality of life (QoL) of children with a diagnosis of cow's milk protein allergy between the ages of 0 and 5 using the Food Allergy Quality of Life Questionnaire - Parent Form. METHODS: A cross-sectional study was conducted in two tertiary care centers in Bogotá. Demographic, medical information, and QoL scores were collected by parental interview. We carried out a cost-of-illness analysis based on self-reported out-of-pocket expenses attributed to the treatment as a whole and the family's monthly income. Exploratory analyses used the QoL scores and the percentage of out-of-pocket expenses attributable to treatment as outcomes. RESULTS: 122 families were analyzed. Median subject age was 17 months (Q1-Q3: 11-26.75 months) and female subjects made up 71% of the sample. The median QoL score was 3.21 points (Q1-Q3: 2.43-4.34) and only differed by age groups and personal history of other food allergies. The median out-of-pocket treatment related costs was 300,000 Colombian pesos (COP) (Q1-Q3: 280,000-340,000 COP). About 17% of the families had to pay over 15% of their monthly income to purchase food and dietary products. Out-of-pocket treatment related costs differed depending on whether the treatment included formulas (Mann-Whitney test p < 0.001). Out-of-pocket treatment expenses were uncorrelated with the QoL scores. CONCLUSION: Food allergy related QoL scores were not associated with out-of-pocket expenses as a whole or as a fraction of monthly income but were higher in children with additional food allergies and in older age groups, suggesting a lower QoL.

Agreement between an Early-Onset Neonatal Sepsis Risk Calculator and the Colombian Clinical Practice Guideline in Three Tertiary-Care Centers in Bogotá, Colombia.

OBJECTIVE: Clinical practice guidelines (CPG) worldwide help steer the management of early-onset neonatal sepsis (EONS). These documents typically discourage the use of risk assessment tools. However, prior work has shown that the Kaiser Permanente calculator (Early-Onset Sepsis Calculator [EOScalc]) could be a useful tool in EONS risk assessment. This study aimed to determine the agreement between the recommendations of the Colombian EONS CPG and those of the EOSCalc tool in a cohort of newborns in Bogotá, Colombia. STUDY DESIGN: Multicenter retrospective observational cohort study. We included newborns with a gestational age ≥ 34 weeks who were admitted to the neonatal care unit with a suspected diagnosis of EONS between 2017 and 2019. Agreement between the two tools was examined using Cohen's kappa under two scenarios (unequivocal and cautious). RESULTS: Of the 23.490 live births, 470 (1.71%) were admitted to the neonatal care unit with a presumptive diagnosis of EONS. This diagnosis was confirmed in seven patients by means of blood cultures, with group B streptococcus the most common organism (57%; 95% confidence interval [CI]: 18.4-90.1). A single death occurred among the patients with confirmed EONS (lethality: 14.3%). The overall incidence of EONS was 0.298 per 1,000 live births. After splitting the recommendations into two scenarios regarding antibiotic use, unequivocal and cautious, the agreement between EOSCalc and the CPG was below 15% (6 and 14%, respectively). CONCLUSION: Recommendations from the Colombian EONS CPG show poor agreement with the EOSCalc, with the latter detecting all newborns with EONS. Although the use of EOSCalc is clinically and administratively advantageous, further prospective studies are warranted to determine the safety of its implementation. KEY POINTS: · Colombian EONS CPGs recommend that an outsized number of newborns be given antibiotics.. · The KP EOSCalc risk assessment calculator shows poor agreement with CPG recommendations.. · The Colombian CPGs should be updated to include the use of risk assessment calculators..

Prediction of patient admission and readmission in adults from a Colombian cohort with bipolar disorder using artificial intelligence.

Introduction: Bipolar disorder (BD) is a chronically progressive mental condition, associated with a reduced quality of life and greater disability. Patient admissions are preventable events with a considerable impact on global functioning and social adjustment. While machine learning (ML) approaches have proven prediction ability in other diseases, little is known about their utility to predict patient admissions in this pathology. Aim: To develop prediction models for hospital admission/readmission within 5 years of diagnosis in patients with BD using ML techniques. Methods: The study utilized data from patients diagnosed with BD in a major healthcare organization in Colombia. Candidate predictors were selected from Electronic Health Records (EHRs) and included sociodemographic and clinical variables. ML algorithms, including Decision Trees, Random Forests, Logistic Regressions, and Support Vector Machines, were used to predict patient admission or readmission. Survival models, including a penalized Cox Model and Random Survival Forest, were used to predict time to admission and first readmission. Model performance was evaluated using accuracy, precision, recall, F1 score, area under the receiver operating characteristic curve (AUC) and concordance index. Results: The admission dataset included 2,726 BD patients, with 354 admissions, while the readmission dataset included 352 patients, with almost half being readmitted. The best-performing model for predicting admission was the Random Forest, with an accuracy score of 0.951 and an AUC of 0.98. The variables with the greatest predictive power in the Recursive Feature Elimination (RFE) importance analysis were the number of psychiatric emergency visits, the number of outpatient follow-up appointments and age. Survival models showed similar results, with the Random Survival Forest performing best, achieving an AUC of 0.95. However, the prediction models for patient readmission had poorer performance, with the Random Forest model being again the best performer but with an AUC below 0.70. Conclusion: ML models, particularly the Random Forest model, outperformed traditional statistical techniques for admission prediction. However, readmission prediction models had poorer performance. This study demonstrates the potential of ML techniques in improving prediction accuracy for BD patient admissions.

Peripheral neuropathy in leprosy: Clinical manifestations and disability in a Colombian national referral center.

To aim of the paper was to describe the neurological features of the physical examination in patients with Hansen's disease who were treated in Bogotá, Colombia. We carried out a descriptive study of all patients with a Hansen's disease diagnosis treated at a referral center between 2003-2018. There were 327 eligible electronic health records (EHRs) with a final sample of 282 subjects. Leprosy was most common in males (57.45%), median age at the diagnosis was 54 years, and lesions of the lower limbs were more common (75.1%). The median time from disease onset to consult was 12 months. Most of them were classified as having lepromatous leprosy (39.7%). Pain over the median nerve trunk was the most common manifestation of disease (28%), followed by pain over the radial trunk (22%). Sensitive alterations were more common than motor ones. Specifically, the posterior tibial nerve was affected in nearly half of subjects. Dual impairment was more common in the ulnar nerve (13.8%). Some disability was apparent in 23.8% of subjects; predominantly grade 1 disability. Findings regarding age, leprosy type, and the frequency of individual nerve compromise were consistent with reports from other countries. Nerve trunk thickening was infrequent, which might be a consequence of subjectiveness in the examination and sample differences in sex distribution, degree of disability and time since disease onset. The frequency of morbidity and disability found in this sample, though low when compared with other series, fails to meet public health goals, including those limiting disability in younger subjects.

Differences Between Methicillin-susceptible Versus Methicillin-resistant Staphylococcus aureus Infections in Pediatrics: Multicenter Cohort Study Conducted in Bogotá, Colombia, 2014-2018.

BACKGROUND: The epidemiology of methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) has changed in recent years. The present article is intended to establish differences between clinical, laboratory and imaging findings and outcomes of MSSA and MRSA infections, as well as among subgroups of infection such as skin and soft tissue infection, osteoarticular, bacteremia or pneumonia in a pediatric population from Bogota, Colombia. METHODS: Retrospective cohort study using clinical records of patients under 18 years of age treated at the participating centers in Bogota, Colombia, between 2014 and 2018. The first positive S. aureus culture was studied. MSSA and MRSA were compared. The χ2 test, Fisher exact test, and Kruskal-Wallis test were calculated, and the statistical significance was presented using the difference and its 95% CI. RESULTS: Five hundred fifty-one patients were included; 211 (38%) corresponded to MRSA and 340 (62%) to MSSA for a total of 703 cultures. A significantly higher probability of having an MSSA infection than MRSA was found in patients with previous heart disease (3.3% vs. 0.5%), neurologic disease (5.9% vs. 2.5%), recent major surgeries (11% vs. 5%) or who has an implanted device (11% vs. 4%). In contrast, in severe MRSA infections (bacteremia, osteoarticular infections and pneumonia), a higher rate of complications was seen (admission to the pediatric intensive care unit, mechanical ventilation and vasoactive support), and in osteoarticular MRSA, more than 1 surgery per case was seen (89% vs. 61%). Laboratory results and mortality were similar. CONCLUSIONS: MRSA was associated with a more severe course in bacteremia, osteoarticular infections and pneumonia. Some classical risk factors associated with MRSA infections were found to be related to MSSA. In general, with the exception of skin and soft tissue infection, there was an increased risk of pediatric intensive care unit admission and mechanical and inotropic support with MRSA in a pediatric population.

Spatial analysis of service areas for stroke centers in a city with high traffic congestion.

The effect that traffic congestion has on the service areas of stroke centers has received scarce attention. We aimed to determine the effect of traffic conditions on the characteristics of service areas of stroke centers in Bogotá, Colombia. Using a webservice, we sampled travel times from a set of census blocks to medical centers offering stroke management in the city. We obtained 179.340 transport times under different conditions. The size of service areas was reduced significantly with congestion (up to 94.83%). Overlap in the locations of centers led to large areas covered by only five centers. We identified areas with transport times to the closest center consistently exceeding 30-minutes to 1-hour in the west and south-west. Traffic conditions in Bogotá significantly affect service areas of centers capable of offering comprehensive stroke care. Spatial overlap of centers led to small catchment areas.