Bjørn Auestad

Nonparametric Identification of Nonlinear Time Series: Projections

Abstract We study the possibility of identifying general linear and nonlinear time series models using nonparametric methods. The kernel estimators of the conditional mean and variance are used as a basis, and the properties of these quantities as model indicators are briefly discussed. Some drawbacks are pointed out, and motivated by these we introduce projections as tools of identification. The projections are especially useful for additive modeling. Expressions for the asymptotic bias and variance are obtained. The projection of the conditional variance is suggested as a tool for identifying heteroscedastic time series. The results are illustrated by simulations for both the estimators of the projections and the estimators of the conditional mean and variance.

Retention of basic life support skills 6 months after training with an automated voice advisory manikin system without instructor involvement

AIM To evaluate the retention of skills 6 months after training in ventilation and chest compressions (CPR) on a manikin with computer based on-line voice advisory feedback and the possible effects of initial overtraining. METHODS Thirty five volunteers had 20 min provisional CPR training on a manikin with computer based voice advisory feedback but without an instructor. The appropriate feedback was taken from a pre-recorded list depending on performance measured by the manikin--computer system versus set limits for ventilation and compression variables. One group in addition was randomised to receive 10 similar 3 min training sessions during 1 week in the following month (overtrained group). All ventilation and compression variables were measured without feedback before and after the initial training session, with feedback immediately thereafter, and both without and with feedback 6 months after the initial training session. RESULTS The initial training improved all variables. Compressions with correct depth increased from a mean of 33 to 77%, and correct inflations from a mean of 9 to 58%. After 6 months, the results for the controls were not significantly different from pre-training, except for a higher of correct inflations (18%), while the overtrained group had better retention of skills including the correct compression depth (mean 61%) and inflations (mean 42%). When verbal feedback was added both the compressions and ventilations immediately improved both when tested immediately and 6 months after the initial training session. CONCLUSIONS The computer-based voice advisory manikin (VAM) feedback system can improve immediate performance of basic life support (BLS) skills, with better long-term retention with overtraining.

Long-Term Follow-Up of the TIPS Early Detection in Psychosis Study: Effects on 10-Year Outcome

OBJECTIVE Early detection in first-episode psychosis confers advantages for negative, cognitive, and depressive symptoms after 1, 2, and 5 years, but longitudinal effects are unknown. The authors investigated the differences in symptoms and recovery after 10 years between regional health care sectors with and without a comprehensive program for the early detection of psychosis. METHOD The authors evaluated 281 patients (early detection, N=141) 18 to 65 years old with a first episode of nonaffective psychosis between 1997 and 2001. Of these, 101 patients in the early-detection area and 73 patients in the usual-detection area were followed up at 10 years, and the authors compared their symptoms and recovery. RESULTS A significantly higher percentage of early-detection patients had recovered at the 10-year follow-up relative to usual-detection patients. This held true despite more severely ill patients dropping out of the study in the usual-detection area. Except for higher levels of excitative symptoms in the early-detection area, there were no symptom differences between the groups. Early-detection recovery rates were higher largely because of higher employment rates for patients in this group. CONCLUSIONS Early detection of first-episode psychosis appears to increase the chances of milder deficits and superior functioning. The mechanisms by which this strategy improves the long-term prognosis of psychosis remain speculative. Nevertheless, our findings over 10 years may indicate that a prognostic link exists between the timing of intervention and outcome that deserves additional study.

Strong and weak aspects of an established post-resuscitation treatment protocol—A five-year observational study

AIM OF STUDY Favourable hospital survival increased from 26% to 56% in the implementation phase of a new standard operating procedure (SOP) for treatment after out-of hospital cardiac arrest (OHCA) in 2003. We now evaluate protocol adherence and survival rates after five years with this established SOP. METHODS This observational study is based on prospectively collected registry data from all OHCA patients with cardiac aetiology admitted with spontaneous circulation to Ulleval Hospital between September 2003 and January 2009. Three patient categories are described based on early assessment in the emergency department: conscious, comatose, and comatose patients receiving only palliative care, with main focus on comatose patients receiving active treatment. RESULTS Of 248 patients, 22% were consciousness on admission, 70% were comatose and received active treatment, while 8% received only palliative care. Favourable survival from admittance to discharge remained at 56% throughout the study period. Among actively treated patients 83% received emergency coronary angiography and 48% underwent subsequent percutaneous coronary intervention. In this cohort 63% had an acute myocardial infarction, ten of whom did not receive emergency coronary angiography. Among actively treated comatose patients, 6% survived with unfavourable neurology, while 51% of the deaths followed treatment withdrawal after prognostication of severe brain injury. CONCLUSION The previously reported doubling in survival rate remained throughout a five-year study period. Establishing reliable indication for emergency coronary angiography and interventions and validating prognostication rules in the hypothermia era are important challenges for future studies.

Nonparametric identification of nonlinear time series: selecting significant lags

Abstract In this article we suggest a nonparametric procedure for selecting significant lags in the model description of a general nonlinear stationary time series. The procedure can be applied to both the conditional mean and the conditional variance and is valid for heteroscedastic series. The procedure is illustrated by simulations and sunspot data, lynx data, and blowfly data are analyzed. It is indicated that projectors can be used in conjunction with the procedure for selecting significant lags to check the adequacy of an additive time series model.

Early detection of psychosis: positive effects on 5-year outcome

BACKGROUND During the last decades we have seen a new focus on early treatment of psychosis. Several reviews have shown that duration of untreated psychosis (DUP) is correlated to better outcome. However, it is still unknown whether early treatment will lead to a better long-term outcome. This study reports the effects of reducing DUP on 5-year course and outcome. METHOD During 1997-2000 a total of 281 consecutive patients aged >17 years with first episode non-affective psychosis were recruited, of which 192 participated in the 5-year follow-up. A comprehensive early detection (ED) programme with public information campaigns and low-threshold psychosis detection teams was established in one healthcare area (ED-area), but not in a comparable area (no-ED area). Both areas ran equivalent treatment programmes during the first 2 years and need-adapted treatment thereafter. RESULTS At the start of treatment, ED-patients had shorter DUP and less symptoms than no-ED-patients. There were no significant differences in treatment (psychotherapy and medication) for the 5 years. Mixed-effects modelling showed better scores for the ED group on the Positive and Negative Syndrome Scale negative, depressive and cognitive factors and for global assessment of functioning for social functioning at 5-year follow-up. The ED group also had more contacts with friends. Regression analysis did not find that these differences could be explained by confounders. CONCLUSIONS Early treatment had positive effects on clinical and functional status at 5-year follow-up in first episode psychosis.

Substance abuse in first-episode non-affective psychosis.

Abuse of alcohol and drugs is an important and clinically challenging aspect of first-episode psychosis. Only a few studies have been carried out on large-sized and reliably characterized samples. These are reviewed, and the results are compared with a sample of 300 first-episode psychosis patients recruited for the TIPS (Early Treatment and Identification of Psychosis) study from Norway and Denmark. Prevalence rates from the literature vary from 6% to 44% for drugs and 3% to 35% for alcohol. In our sample, 23% abused drugs and 15% abused alcohol during the last 6 months. When compared to non-abusers, the drug-abusing group is characterized by the following: male gender, younger age, better premorbid social, poor premorbid academic functioning, and more contact with friends in the last year before onset. Alcohol abusers were the oldest group and they had the least contact with friends. A group of patients abusing both drugs and alcohol had poor premorbid academic functioning from early childhood. Overall, drug and alcohol abuse are highly prevalent in contemporary first-episode psychosis samples. In our study, substance abuse comorbidity did not generate differences on diagnosis, duration of untreated psychosis, psychiatric symptoms, or global functioning at onset/baseline. The premorbid profiles of the substance abusers were clearly different from the non-abusers. Drug abusers, in particular, were more socially active both premorbidly and during the year preceding the start of treatment.

Efficacy of memantine for agitation in Alzheimer's dementia: a randomised double-blind placebo controlled trial.

Background Agitation in Alzheimer’s disease (AD) is common and associated with poor patient life-quality and carer distress. The best evidence-based pharmacological treatments are antipsychotics which have limited benefits with increased morbidity and mortality. There are no memantine trials in clinically significant agitation but post-hoc analyses in other populations found reduced agitation. We tested the primary hypothesis, memantine is superior to placebo for clinically significant agitation, in patients with moderate-to-severe AD. Methods and Findings We recruited 153 participants with AD and clinically significant agitation from care-homes or hospitals for a double-blind randomised-controlled trial and 149 people started the trial of memantine versus placebo. The primary outcome was 6 weeks mixed model autoregressive analysis of Cohen-Mansfield Agitation Inventory (CMAI). Secondary outcomes were: 12 weeks CMAI; 6 and 12 weeks Neuropsychiatric symptoms (NPI), Clinical Global Impression Change (CGI-C), Standardised Mini Mental State Examination, Severe Impairment Battery. Using a mixed effects model we found no significant differences in the primary outcome, 6 weeks CMAI, between memantine and placebo (memantine lower −3.0; −8.3 to 2.2, p = 0.26); or 12 weeks CMAI; or CGI-C or adverse events at 6 or 12 weeks. NPI mean difference favoured memantine at weeks 6 (−6.9; −12.2 to −1.6; p = 0.012) and 12 (−9.6; −15.0 to −4.3 p = 0.0005). Memantine was significantly better than placebo for cognition. The main study limitation is that it still remains to be determined whether memantine has a role in milder agitation in AD. Conclusions Memantine did not improve significant agitation in people with in moderate-to-severe AD. Future studies are urgently needed to test other pharmacological candidates in this group and memantine for neuropsychiatric symptoms. Trial Registration ClinicalTrials.gov NCT00371059 Trial Registration International Standard Randomised Controlled Trial 24953404

Delaying Defibrillation to Give Basic Cardiopulmonary Resuscitation to Patients With Out-of-Hospital Ventricular Fibrillation

Context Defibrillationassoonaspossibleisstandardtreatmentforpatientswithven-tricular fibrillation. A nonrandomized study indicates that after a few minutes of ven-tricular fibrillation, delaying defibrillation to give cardiopulmonary resuscitation (CPR)first might improve the outcome.Objective To determine the effects of CPR before defibrillation on outcome in pa-tients with ventricular fibrillation and with response times either up to or longer than5 minutes.Design, Setting, and Patients Randomized trial of 200 patients with out-of-hospital ventricular fibrillation in Oslo, Norway, between June 1998 and May 2001.Patientsreceivedeitherstandardcarewithimmediatedefibrillation(n=96)orCPRfirstwith 3 minutes of basic CPR by ambulance personnel prior to defibrillation (n=104).If initial defibrillation was unsuccessful, the standard group received 1 minute of CPRbeforeadditionaldefibrillationattemptscomparedwith3minutesintheCPRfirstgroup.MainOutcomeMeasure Primaryendpointwassurvivaltohospitaldischarge.Sec-ondaryendpointswerehospitaladmissionwithreturnofspontaneouscirculation(ROSC),1-year survival, and neurological outcome. A prespecified analysis examined sub-groups with response times either up to or longer than 5 minutes.Results In the standard group, 14 (15%) of 96 patients survived to hospital dis-charge vs 23 (22%) of 104 in the CPR first group (

Neonatal Outcome Following Cord Clamping After Onset of Spontaneous Respiration

BACKGROUND AND OBJECTIVE: Evolving data indicate that cord clamping (CC) beyond 30 to 60 seconds after birth is of benefit for all infants. Recent experimental data demonstrated that ventilation before CC improved cardiovascular stability by increasing pulmonary blood flow. The objective was to describe the relationship between time to CC, onset of spontaneous respirations (SR), and 24-hour neonatal outcome. METHODS: In a rural Tanzanian hospital, trained research assistants, working in shifts, have observed every delivery (November 2009–February 2013) and recorded data including time interval from birth to SR and CC, fetal heart rate, perinatal characteristics and outcome (normal, death, admission). RESULTS: Of 15 563 infants born, 12 780 (84.3%) initiated SR at 10.8 ± 16.7 seconds, and CC occurred at 63 ± 45 seconds after birth. Outcomes included 12 730 (99.7%) normal, 31 deaths, and 19 admitted; 11 967 were of birth weight (BW) ≥2500 g and 813 <2500 g. By logistic modeling, the risk of death/admission was consistently higher if CC occurred before SR. Infants of BW <2500 g were more likely to die or be admitted. The risk of death/admission decreased by 20% for every 10-second delay in CC after SR; this risk declined at the same rate in both BW groups. CONCLUSIONS: Healthy self-breathing neonates are more likely to die or be admitted if CC occurs before or immediately after onset of SR. These clinical observations support the experimental findings of a smoother cardiovascular transition when CC is performed after initiation of ventilation.