Trends in Biochemical Parameters, Healthcare Resource and Medication Use in the 5 Years Preceding IBD Diagnosis: A Health Maintenance Organization Cohort Study.

BACKGROUND: Few data describing pre-diagnosis changes in patients with inflammatory bowel disease (IBD) exist. We aimed to determine if there is a pattern of change in use of health resources, medications and laboratory results in the years preceding diagnosis. METHODS: This retrospective study used electronic medical records of Maccabi Health Services (MHS). Patients with IBD ≥ 16 years of age and minimum of 5-years follow-up were identified by entry into the MHS IBD registry and included in the analysis. Demographic, clinical, medication and laboratory data were collected. Generalized estimating equation model was applied to study trends and compare between years. RESULTS: This study included 5643 patients with IBD. Of these, 3039 (53.8%) had Crohn's disease (CD), 2322 (41.1%) had ulcerative colitis (UC) and 282 (5%) had indeterminate colitis (IC). Laboratory parameters including white blood cells, platelets and C-reactive protein showed significant increases while haemoglobin and mean cell volume showed significant decreases in mean values in the 2 years prior to diagnosis with stable values prior to that (p < 0.0001). Parameters such as creatinine, total protein and albumin showed significant, progressive decreases in mean values starting 5 years prior to diagnosis (p < 0.0001). Patients with CD had distinct laboratory trends when compared with patients with UC. CONCLUSIONS: Changes in laboratory parameters, healthcare service and medication use occur during the 5-year period before IBD diagnosis. These data can have future clinical applicability by developing a composite score and referral algorithm introducing red flags into primary care visits and appropriate referral for specialist care.

Temporal patterns selection for All-Cause Mortality prediction in T2D with ANNs.

Mortality prevention in T2D elderly population having Chronic Kidney Disease (CKD) may be possible thorough risk assessment and predictive modeling. In this study we investigate the ability to predict mortality using heterogeneous Electronic Health Records data. Temporal abstraction is employed to transform the heterogeneous multivariate temporal data into a uniform representation of symbolic time intervals, from which then frequent Time Intervals Related Patterns (TIRPs) are discovered. However, in this study a novel representation of the TIRPs is introduced, which enables to incorporate them in Deep Learning Networks. We describe here the use of iTirps and bTirps, in which the TIRPs are represented by a integer and binary vector representing the time respectively. While bTirp represents whether a TIRP's instance was present, iTirp represents whether multiple instances were present. While the framework showed encouraging results, a major challenge is often the large number of TIRPs, which may cause the models to under-perform. We introduce a novel method for TIRPs' selection method, called TIRP Ranking Criteria (TRC), which is consists on the TIRP's metrics, such as the differences in its recurrences, its frequencies, and the average duration difference between the classes. Additionally, we introduce an advanced version, called TRC Redundant TIRP Removal (TRC-RTR), TIRPs that highly correlate are candidates for removal. Then the selected subset of iTirp/bTirps is fed into a Deep Learning architecture like a Recurrent Neural Network or a Convolutional Neural Network. Furthermore, a predictive committee is utilized in which raw data and iTirp data are both used as input. Our results show that iTirps-based models that use a subset of iTirps based on the TRC-RTR method outperform models that use raw data or models that use full set of discovered iTirps.

All-cause mortality prediction in T2D patients with iTirps.

Mortality in the type II diabetic elderly population can sometimes be prevented through intervention, for which risk assessment through predictive modeling is required. Since Electronic Health Records data are typically heterogeneous and sparse, the use of Temporal Abstraction and time intervals mining to discover frequent Time Intervals Related Patterns (TIRPs) is employed. While TIRPs are used as features for a predictive model, the temporal relations between them in general, and among each TIRP's instances are not represented. We introduce a novel TIRP based representation called integer-TIRP (iTirp) in which the TIRPs become channels containing values that represent the TIRP instances that were detected at each time point. Then the iTirp representation is fed into a Deep Learning architecture, that learns this kind of temporal relations, using a Recurrent Neural Network or a Convolutional Neural Network. Additionally, a predictive committee is introduced in which raw data and iTirp data are concatenated as inputs. Our results show that iTirps based models outperform the use of deep learning with raw data, resulting in 82% AUC.

The implementation of clinical guidelines concerning return to work after myocardial infarction.

PURPOSE: To examine the implementation, clinical effects, and adherence of occupational physicians to the clinical practice guidelines (CPG) published in 2003 in Israel, regarding return to work following acute myocardial infarction (AMI). METHODS: We analyzed 3 populations referred to Maccabi Health Services occupational clinics for fitness for work examinations following AMI, either before the guidelines were published (group 1), after their first publications in 2003 (group 2) or after they were re-published in Harefuah Journal in 2007 (group3). RESULTS: The documentation among groups 1,2,3 of different parameters like years of education (18.8%, 45.1%. 57.5%, p < 0.001), type of occupation (86.5%, 98.2%, 100.0%, p < 0.001) and New York Heart Association (NYHA) class (69.8%, 81.4%, 92.5%, p < 0.001) has all improved respectively over the years. No significant difference in sick leave was noted between the three populations. CONCLUSION: The CPG should be interdisciplinary in the phase of writing and publication. Economical resources are needed in order to implement CPG.Implications for RehabilitationThe clinical practice guidelines should be interdisciplinary in the phase of writing and should involve all rehabilitation professionals as a team.The clinical practice guidelines need disseminating widely alongside a resource plan in order to implement it.

Urinary tract infections in children < 2 years of age hospitalized in a tertiary medical center in Southern Israel: epidemiologic, imaging, and microbiologic characteristics of first episode in life.

The objectives of the study were to describe the epidemiologic, microbiologic, and imaging characteristics of first 44 UTI episode in hospitalized infants and children < 2 years of age. A UTI episode was diagnosed based on suprapubic aspiration or urinary catheterization and report of a significant bacterial growth of true uropathogens. Two thousand two hundred ninety-four UTI episodes were recorded during 2009-2013 in 1694 patients; 1350 (79.7%), 223 (13.2%), and 66 (3.9%) had one, two, and three episodes, respectively. Of 1955 pathogens isolated, the most frequent were E. coli, Klebsiella spp., and Enterococcus spp. (56.9%, 14.1%, and 11%, respectively). E. coli percentages increased with increase in patient age while Klebsiella spp. and Enterococcus spp. decreased with decrease in age. 136/344 (39.5%) renal ultrasound examinations performed were reported abnormal. The percentages of abnormal ultrasound examinations in Enterococcus spp.-UTI were higher than in E. coli and Klebsiella spp.-UTI (P < 0.001 and P = 0.007, respectively). The E. coli nonsusceptibility to ampicillin, TMP/SMX, ceftriaxone, amoxicillin/clavulanic acid, cefuroxime, and gentamicin was 71%, 31%, 14.9%, 14.7%, 7%, and 4.4%, respectively. Nonsusceptibility of Klebsiella spp. to ampicillin, amoxicillin/clavulanic acid, TMP/SMX, ceftriaxone, gentamicin, and cefuroxime was 98.7%, 10.4%, 9.5%, 8.4%, 4.8%, and 4.3%, respectively. E. coli nonsusceptibility to amoxicillin/clavulanate, TMP/SMX, and ciprofloxacin decreased during the study period. E. coli percentages increased and those of Klebsiella spp. and Enterococcus spp. decreased with increase in patient age. Enterococcus. spp.-UTI was associated with older age and more severe findings on renal ultrasound. E. coli and Klebsiella spp. nonsusceptibility to amoxicillin/clavulanate, TMP/SMX, cefuroxime, and ceftriaxone was low.

Epidemiologic, clinical, laboratory, and therapeutic characteristics of influenza A/H1N1 in Moslem Bedouin and Jewish children hospitalized in southern Israel during 2009.

A total of 739 (225 H1N1(+)) children with a diagnosis of acute respiratory infection were hospitalized during July to December 2009. The H1N1(+) children were compared with 225 randomly enrolled H1N1(-) children with an influenza-like illness. As compared with influenza-like illness patients, patients with 2009 influenza A/H1N1 were characterized by older age, more vomiting, less hypoxemia and wheezing, lower white blood cell counts, less neutrophilia, and severe lymphopenia.