Comparison of machine learning algorithms for clinical event prediction (risk of coronary heart disease).

AIM: The aim of this study is to compare the utility of several supervised machine learning (ML) algorithms for predicting clinical events in terms of their internal validity and accuracy. The results, which were obtained using two statistical software platforms, were also compared. MATERIALS AND METHODS: The data used in this research come from the open database of the Framingham Heart Study, which originated in 1948 in Framingham, Massachusetts as a prospective study of risk factors for cardiovascular disease. Through data mining processes, three data models were elaborated and a comparative methodological study between the different ML algorithms - decision tree, random forest, support vector machines, neural networks, and logistic regression - was carried out. The global selection criterium for choosing the right set of hyperparameters and the type of data manipulation was the area under a curve (AUC). The software tools used to analyze the data were R-Studio® and RapidMiner®. RESULTS: The Framingham study open database contains 4240 observations. The algorithm that yielded the greatest AUC when analyzing the data in R-Studio was neural network applied to a model that excluded all observations in which there was at least one missing value (AUC = 0.71); when analyzing the data in RapidMiner and applying the same model, the best algorithm was support vector machines (AUC = 0.75). CONCLUSIONS: ML algorithms can reinforce the diagnostic and prognostic capacity of traditional regression techniques. Differences between the applicability of those algorithms and the results obtained with them were a function of the software platforms used in the data analysis.

Perceptions of co-teaching as a pedagogical approach to integrate basic and clinical sciences.

Background: Medical curricula must provide students with basic and clinical competencies for critical reasoning and diagnosing. These competencies are better acquired when basic and clinical science are taught in an integrated and collaborative manner. In this study, we investigate whether supportive co-teaching (SCT) is an effective approach to promote integrated and reasoned learning as well as to help medical students applying theoretical concepts to clinical scenarios taught in a team-based learning (TBL) framework. Methods: We conducted a concurrent mixed methods study. For the qualitative part, we performed a focus group and semi-structured interviews to clinical and basic science teachers and medical students. Using conventional content analysis, themes were identified deductively. For the quantitative part, an analytical and descriptive observational study of the 2019-2020 cohort of first-year undergraduate medical students was conducted (107 students out of 220 completed the survey). For the descriptive study, questions were grouped into 5 categories. Results: Deductive themes from the analysis include relationship between clinical and basic science teachers, knowledge integration, methodology, teamwork and integrated Medicine and curricular design. Basic science and clinical teachers highlighted their relationship as critical to increase their mutual knowledge. This was supported by the student's opinion who very much valued their joint feedback. Regarding knowledge integration, both teachers and students found that horizontal and vertical integration enhanced applicability of basic knowledge to future clinical practice. The TBL methodology was very well perceived by both students and teachers and was highly motivating for students even though the need for commitment. Students considered that this program presented a great opportunity and expressed their interest in maintaining it in the future. These results were supported by the quantitative data. Conclusion: Our work supports the value of co-teaching in basic and clinical sciences within a TBL framework set in real clinical case scenarios. By employing this approach, students can actively apply their theoretical knowledge to clinical practice, enhancing their critical thinking, problem-solving, and clinical reasoning skills. Our findings can inform curriculum design and improved educational practice, leading to enhanced learning experiences for healthcare students and ultimately better patient care.

[HIV infection in the adult: epidemiological changes over 25 years (1983-2008) in an area of the Community of Madrid]. / Infección por el VIH en el adulto: cambios epidemiológicos durante 25 años (1983-2008) en un área de la Comunidad de Madrid.

OBJECTIVE: To describe the epidemiological characteristics and evolution of a cohort of HIV-infected patients in Madrid (Spain) over a period of 25 years. METHODS: Longitudinal, prospective, cohort study of all patients diagnosed with HIV infection seen at the Hospital de Móstoles (1983-2008). RESULTS: Of the 2156 patients attended, 73% were men. In 68%, the route of infection was illicit drug use. There was a peak of new diagnoses in 1991 (188 patients) and a subsequent gradual decline. Sexual transmission increased over the time period studied, with a higher rise in heterosexual transmission. The percentage of immigrants with the infection also increased (more than 60% in the last 3 years). Sub-Saharan Africa remains the predominant region of origin of our immigrant patients. Over the period studied, 5% of patients were diagnosed during the acute infection, and 20% to 30% were diagnosed with AIDS within 1 year after detection of HIV infection. The estimated rate of new diagnoses of HIV infection in 2006 was 195 per million population (pmp), 269 pmp in men and 121 pmp in women. CONCLUSIONS: Many changes have occurred in the epidemiology of HIV infection in our area in the last 25 years. Routine serologic testing without waiting for warning signs would have an impact on early diagnosis and improving the prognosis of these patients.

Glucose series complexity in hypertensive patients.

Nonlinear methods have been applied to the analysis of biological signals. Complexity analysis of glucose time series may be a useful tool for the study of the initial phases of glucoregulatory dysfunction. This observational, cross-sectional study was performed in patients with essential hypertension. Glucose complexity was measured with detrended fluctuation analysis (DFA), and glucose variability was measured by the mean amplitudes of glycemic excursion (MAGE). We included 91 patients with a mean age of 59 ± 10 years. We found significant correlations for the number of metabolic syndrome (MS)-defining criteria with DFA (r = 0.233, P = .026) and MAGE (r = 0.396, P < .0001). DFA differed significantly between patients who complied with MS and those who did not (1.44 vs. 1.39, P = .018). The MAGE (f = 5.3, P = .006), diastolic blood pressures (f = 4.1, P = .018), and homeostasis model assessment indices (f = 4.2, P = .018) differed between the DFA tertiles. Multivariate analysis revealed that the only independent determinants of the DFA values were MAGE (ß coefficient = 0.002, 95% confidence interval: 0.001-0.004, P = .001) and abdominal circumference (ß coefficient = 0.002, 95% confidence interval: 0.000015-0.004, P = .048). In our population, DFA was associated with MS and a number of MS criteria. Complexity analysis seemed to be capable of detecting differences in variables that are arguably related to the risk of the development of type 2 diabetes.

Cystatin C as a predictor of mortality and cardiovascular events in a population with chronic kidney disease.

Background. We examine whether cystatin C, a surrogate marker of renal function, could identify patients with chronic kidney disease (CKD) with an increased risk of renal disease progression, death, or cardiovascular events. Methods. Data were obtained for 180 patients, with a diagnosis of chronic renal failure based on serum creatinine estimated glomerular filtration rate (eGFRcreat) <90 mL/min/1.73 m(2). This population was grouped in tertiles according to cystatin C and creatinine values at baseline. Cardiovascular events and overall mortality were estimated for each tertile. Predictors of overall mortality and for the development of renal disease progression were analyzed. Results. The median age was 75 years (interquartile range 69-82) and the median eGFRcreat 38 mL/min m(2) (interquartile range 33-49). Overall mortality was lower on the first and on the second tertiles of cystatin C than on the third one (HR = 0.060; 95% CI: 0.008-0.447 and HR = 0.094; 95% CI: 0.022-0.406, resp.). Deaths related to the creatinine tertiles followed the same pattern, but differences were not as large. Cardiovascular mortality was lower on the second than on the third cystatin C tertile (HR = 0.198; 95% CI: 0.040-0.987), but it did not show differences on the first and the second creatinine tertiles compared with the third one (HR = 0.126; 95% CI: 0.013-1.265 and HR = 0.403; 95% CI: 0.093-1.740). The only independent predictors of mortality during followup were baseline cystatin C (OR = 0.100; 95% CI: 0.021-0.463) and baseline uric acid (OR = 1.377; 95% CI: 1.070-1.773). Conclusion. Cystatin C may be an alternative to creatinine for detecting a high risk of death and cardiovascular events in a population with CKD.

Symptomatic and asymptomatic interphalageal osteoarthritis: An ultrasonographic study.

OBJECTIVE: To date few studies have examined whether ultrasonography can depict morphologic differences in painful and painless osteoarthritis (OA). This study describes and compares the clinical, radiographic and ultrasonographic findings of patients with both painful and painless proximal interphalgeal (PIP) and/or distal interphalgeal (DIP) OA. METHODS: Patients with PIP and/or DIP OA (ACR criteria) were prospectively recruited. The clinical rheumatologist chose up to 3 painful joints and up to 3 painless symmetric joints in each patient to define 2 cohorts of OA: symptomatic (SG) and asymptomatic (ASG). A conventional postero-anterior hand x ray was performed and read by one rheumatologist following the OARSI atlas, blinded to clinical and sonographic data. Ultrasound (US) was performed by an experienced rheumatologist, blinded to both clinical and radiographic data in joints previously selected by the clinical rheumatologist. US-pathology was assessed as present or absent as defined in previous reports: osteophytes, joint space narrowing, synovitis, intra-articular power doppler signal, intra-articular bony erosion, and visualization of cartilage. Radiographic and ultrasonographic intrareader reliability test was performed. RESULTS: A total of 50 joints in the SG and ASG were included from 20 right handed women aged 61.85 (46-73) years with PIP and DIP OA diagnosed 6.8 (1-17) years ago. 70% SG joints and ASG were right and left sided respectively. The SG showed significantly more osteophytes, synovitis and non-visualization of joint cartilage. Intrareader radiographic and ultrasonographic agreement was excellent. CONCLUSION: This study demonstrates that painful PIP and/or DIP OA have more ultrasonographic structural changes and synovitis.

Cystatin C is associated with the metabolic syndrome and other cardiovascular risk factors in a hypertensive population.

Serum cystatin C has been associated with cardiovascular disease. We investigated whether cystatin C concentration is associated with the metabolic syndrome and with other cardiovascular risk factors in a hypertensive population. In this cross-sectional study, we prospectively included 611 essential hypertensive patients during a 12-month period. Cystatin C concentration was measured by nephelometry. The metabolic syndrome was present in 46% of the patients. Cystatin C was significantly higher in patients with the metabolic syndrome (0.94 +/- 0.27 mg/L) than in those without (0.87 +/- 0.23 mg/L) (P < .0001). Pearson partial correlation analysis showed a significant correlation between cystatin C and body mass index (r = 0.240; P = .001); waist circumference (r = 0.173; P = .012); microalbuminuria (r = 0.273; P < .0001); triglycerides (r = 0.138; P = .047); C-reactive protein (r = 0.190; P = .006); uric acid (r = 0.284; P < .0001); age (r = 0.409; P < .0001); and glomerular filtration rate (GFR) (r = -0.638; P < .0001). Multivariate analysis showed that GFR (B = -0.0061; 95% confidence interval [CI], -0.0073 to -0.0049; P < .0001), age (B = 0.0023; 95% CI, 0.0005-0.0041; P = .009), microalbuminuria (B = 0.0005; 95% CI, 0.0002-0.0007; P < .0001), uric acid (B = 0.0252; 95% CI, 0.0085-0.0418; P = .003), body mass index (B = 0.0051, 95% CI, 0.0012-0.0089; P = .011), and C-reactive protein (B = 0.0048; 95% CI, 0.0015-0.0082; P = .005) were independent determinants of cystatin C concentration. Measuring cystatin C concentration in hypertensive patients may be useful for evaluating their cardiovascular risk profile.

Effect of simultaneous use of highly active antiretroviral therapy on survival of HIV patients with tuberculosis.

INTRODUCTION: The optimal timing for initiation of highly active antiretroviral therapy (HAART) in patients with AIDS and tuberculosis (TB) is an unresolved question. To assess the effect of HAART on the survival of patients with TB, we designed this study. METHODS: We selected all HIV patients included in the COMESEM cohort with TB diagnosis after 1996. Clinical and epidemiological data were registered. We compared patients who started HAART at the diagnosis of TB [simultaneous therapy (ST)] or not. Survival was assessed by Cox analysis. RESULTS: Among the 6934 HIV patients included in the cohort, 1217 patients had TB, 322 of them (26.5%) after 1996. At the time of TB diagnosis, 45% of them started HAART (ST). There were no differences between groups regarding basal characteristics, except for a lower viral load in ST patients. ST therapy was associated with improved survival (hazard ratio 0.38; 95% confidence interval 0.20 to 0.72, P = 0.003). By univariate analysis, survival was also associated with no endovenous drug use and a later year of TB diagnosis. After adjusting for other prognostic variables, by Cox multivariate analysis, ST remained robustly associated with improved survival (hazard ratio 0.37; 95% confidence interval 0.17 to 0.66, P = 0.001). CONCLUSIONS: Simultaneous HAART and TB treatment in HIV patients with TB is associated with improved survival.

[Validation of a questionnaire to estimate satisfaction with antiretroviral treatment: CESTA questionnaire]. / Validación del cuestionario de satisfacción con el tratamiento antirretroviral: cuestionario CESTA.

BACKGROUND: Assessment of patient satisfaction with antiretroviral therapy is generating increasing interest in clinical practice, since the outcome is directly related with compliance with therapy and its effectiveness. Currently, there is no validated patient satisfaction questionnaire to evaluate this factor as related to control of the disease. PATIENTS AND METHODS: An "ad hoc" questionnaire--Cuestionario Español de Satisfacción con el Tratamiento Antiretroviral (CESTA), Spanish Antiretroviral Treatment Satisfaction Questionnaire--was developed to evaluate satisfaction in patients switching to a simplified regimen. In a second phase, the questionnaire was validated in an observational study including 321 patients divided into two cohorts (patients switching to a simplified regimen and patients maintaining the same regimen). RESULTS: A total of 99% and 93% patients, respectively, completed the questionnaire at baseline and at the end-of-study visit. Cronbach's alpha coefficient for internal consistency and test-retest reliability were 0.82 and 0.69, respectively. Patients switching to a simplified regimen showed a statistically significant higher overall satisfaction score at the third-month visit than at baseline. CONCLUSIONS: The CESTA questionnaire is a valid instrument for use in clinical practice and provides valuable information on patient satisfaction with antiretroviral therapy.

Lipid disorders in antiretroviral-naive patients treated with lopinavir/ritonavir-based HAART: frequency, characterization and risk factors.

OBJECTIVES: The aim of this study was to evaluate the frequency, characteristics and risk factors of lipid changes associated with lopinavir/ritonavir treatment in antiretroviral-naive patients. METHODS: A prospective cohort of 107 antiretroviral-naive HIV-infected patients was followed for 12 months after starting lopinavir/ritonavir-based highly active antiretroviral therapy. RESULTS: At 12 months, percentages of patients with hypercholesterolaemia and hypertriglyceridaemia were 17.4% and 40%, respectively. Mean increases in total cholesterol and triglycerides were 40.7 and 73.3 mg/dL. There was a significant increase in both low-density and high-density (HDL) cholesterol, and no increase in the total cholesterol/HDL ratio (from 4.16 at baseline to 4.49 after 12 months). Baseline cholesterol > 200 mg/dL and triglycerides > 150 mg/dL were independent risk factors for dyslipidaemia, while hepatitis C coinfection appeared to be protective. CONCLUSIONS: Patients with elevated lipid values at baseline have the greatest risk of developing hypercholesterolaemia and hypertriglyceridaemia after starting lopinavir/ritonavir. Antiretroviral-naive patients coinfected with hepatitis C have a low risk of developing hyperlipidaemia after starting lopinavir/ritonavir.