Prediction of postpartum depression using multilayer perceptrons and pruning.

OBJECTIVE. The main goal of this paper is to obtain a classification model based on feed-forward multilayer perceptrons in order to improve postpartum depression prediction during the 32 weeks after childbirth with a high sensitivity and specificity and to develop a tool to be integrated in a decision support system for clinicians. MATERIALS AND METHODS. Multilayer perceptrons were trained on data from 1397 women who had just given birth, from seven Spanish general hospitals, including clinical, environmental and genetic variables. A prospective cohort study was made just after delivery, at 8 weeks and at 32 weeks after delivery. The models were evaluated with the geometric mean of accuracies using a hold-out strategy. RESULTS. Multilayer perceptrons showed good performance (high sensitivity and specificity) as predictive models for postpartum depression. CONCLUSIONS. The use of these models in a decision support system can be clinically evaluated in future work. The analysis of the models by pruning leads to a qualitative interpretation of the influence of each variable in the interest of clinical protocols.

Study protocol for a pragmatic randomised controlled trial in general practice investigating the effectiveness of acupuncture against migraine.

BACKGROUND Migraine is a chronic neurologic disease that can severely affect the patient's quality of life. Although in recent years many randomised studies have been carried out to investigate the effectiveness of acupuncture as a treatment for migraine, it remains a controversial issue. Our aim is to determine whether acupuncture, applied under real conditions of clinical practice in the area of primary healthcare, is more effective than conventional treatment. METHODS/DESIGN The design consists of a pragmatic multi-centre, three-armed randomised controlled trial, complemented with an economic evaluation of the results achieved, comparing the effectiveness of verum acupuncture with sham acupuncture, and with a control group receiving normal care only. Patients eligible for inclusion will be those presenting in general practice with migraine and for whom their General Practitioner (GP) is considering referral for acupuncture. Sampling will be by consecutive selection, and by randomised allocation to the three branches of the study, in a centralised way following a 1:1:1 distribution (verum acupuncture; sham acupuncture; conventional treatment). Secondly, one patient in three will be randomly selected from each of the acupuncture (verum or sham) groups for a brain perfusion study (by single photon emission tomography). The treatment with verum acupuncture will consist of 8 treatment sessions, once a week, at points selected individually by the acupuncturist. The sham acupuncture group will receive 8 sessions, one per week, with treatment being applied at non-acupuncture points in the dorsal and lumbar regions, using the minimal puncture technique. The control group will be given conventional treatment, as will the other two groups. DISCUSSION This trial will contribute to available evidence on acupuncture for the treatment of migraine. The primary endpoint is the difference in the number of days with migraine among the three groups, between the baseline period (the 4 weeks prior to the start of treatment) and the period from weeks 9 to 12. As a secondary aspect, we shall record the index of laterality and the percentage of change in the mean count per pixel in each region of interest measured by the brain perfusion tomography, performed on a subsample of the patients within the real and sham acupuncture groups. TRIAL REGISTRATION Current Controlled Trials ISRCTN98703707.

Fosfomycin versus meropenem in bacteraemic urinary tract infections caused by extended-spectrum β-lactamase-producing Escherichia coli (FOREST): study protocol for an investigator-driven randomised controlled trial.

INTRODUCTION Finding therapeutic alternatives to carbapenems in infections caused by extended-spectrum β-lactamase-producing Escherichia coli (ESBL-EC) is imperative. Although fosfomycin was discovered more than 40 years ago, it was not investigated in accordance with current standards and so is not used in clinical practice except in desperate situations. It is one of the so-called neglected antibiotics of high potential interest for the future. METHODS AND ANALYSIS The main objective of this project is to demonstrate the clinical non-inferiority of intravenous fosfomycin with regard to meropenem for treating bacteraemic urinary tract infections (UTI) caused by ESBL-EC. This is a 'real practice' multicentre, open-label, phase III randomised controlled trial, designed to compare the clinical and microbiological efficacy, and safety of intravenous fosfomycin (4 g/6 h) and meropenem (1 g/8 h) as targeted therapy for this infection; a change to oral therapy is permitted after 5 days in both arms, in accordance with predetermined options. The study design follows the latest recommendations for designing trials investigating new options for multidrug-resistant bacteria. Secondary objectives include the study of fosfomycin concentrations in plasma and the impact of both drugs on intestinal colonisation by multidrug-resistant Gram-negative bacilli. ETHICS AND DISSEMINATION Ethical approval was obtained from the Andalusian Coordinating Institutional Review Board (IRB) for Biomedical Research (Referral Ethics Committee), which obtained approval from the local ethics committees at all participating sites in Spain (22 sites). Data will be presented at international conferences and published in peer-reviewed journals. DISCUSSION This project is proposed as an initial step in the investigation of an orphan antimicrobial of low cost with high potential as a therapeutic alternative in common infections such as UTI in selected patients. These results may have a major impact on the use of antibiotics and the development of new projects with this drug, whether as monotherapy or combination therapy. TRIAL REGISTRATION NUMBER NCT02142751. EudraCT no: 2013-002922-21. Protocol V.1.1 dated 14 March 2014.