OHDSI MEETINGS THIS WEEK
Patient-level prediction (Western hemisphere) workgroup meeting - Wednesday at 12pm ET
Population-Level Estimation (Eastern hemisphere) WG Meeting - Wednesday at 3pm Hong Kong time
Architecture WG meeting - Thursday at 1pm ET
Webex: https://jjconferencing.webex.com/mw3000/mywebex/default.do?service=1&main_url=%2Fmc3000%2Fe.do%3Fsiteurl%3Djjconferencing%26AT%3DMI%26EventID%3D283835502%26MTID%3Dmb7e839a762fbdaab0608f27500679223%26Host%3DQUhTSwAAAASTOYBxx2KWYTIL7ZSgHgHJoIHctjgxp8k5mgqAEPo2a1ESa12-8hjAVWbZznuDt96ugkT31G8sY5iSWLRwSwhI0%26FrameSet%3D2&siteurl=jjconferencing&nomenu=true ANNOUNCEMENTS
Hadoop Hack-A-Thon - Our first Hadoop Hack-A-Thon is taking place this week (June 28-29th) at QuintilesIMS. Details, including travel and accommodation, are available here: https://www.ohdsi.org/events/hadoop-hack-a-thon/
Call for Sponsorship - To fund the 2017 symposium we need to raise $200K to cover operating costs of venue rental, lunch, audio/visuals and recordings fees. We’re reaching out to the entire community for sponsors who want to support this important event and allow us to build upon the success of past symposia. More details: OHDSI Symposium 2017 - call for sponsorship
The secret of getting ahead is getting started.
Predicting Biomedical Metadata in CEDAR: a Study of Gene Expression Omnibus (GEO).
M Panahiazar, M Dumontier and O Gevaert,
Journal of biomedical informatics, 2017 08
A crucial and limiting factor in data reuse is the lack of accurate, structured, and complete descriptions of data, known as metadata. Towards improving the quantity and quality of metadata, we propose a novel metadata prediction framework to learn associations from existing metadata that can be used to predict metadata values. We evaluate our framework in the context of experimental metadata from the Gene Expression Omnibus (GEO). We applied four rule mining algorithms to the most common structured metadata elements (sample type, molecular type, platform, label type and organism) from over 1.3million GEO records. We examined the quality of well supported rules from each algorithm and visualized the dependencies among metadata elements. Finally, we evaluated the performance of the algorithms in terms of accuracy, precision, recall, and F-measure. We found that PART is the best algorithm outperforming Apriori, Predictive Apriori, and Decision Table. All algorithms perform significantly better in predicting class values than the majority vote classifier. We found that the performance of the algorithms is related to the dimensionality of the GEO elements. The average performance of all algorithm increases due of the decreasing of dimensionality of the unique values of these elements (2697 platforms, 537 organisms, 454 labels, 9 molecules, and 5 types). Our work suggests that experimental metadata such as present in GEO can be accurately predicted using rule mining algorithms. Our work has implications for both prospective and retrospective augmentation of metadata quality, which are geared towards making data easier to find and reuse.
Antibiotic Use in Children - A Cross-National Analysis of 6 Countries.
I Youngster, J Avorn, V Belleudi, A Cantarutti, J Díez-Domingo, U Kirchmayer, BJ Park, S Peiró, G Sanfélix-Gimeno, H Schröder, K Schüssel, JY Shin, SM Shin, GS Simonsen, HS Blix, A Tong, G Trifirò, T Ziv-Baran and SC Kim,
The Journal of pediatrics, 2017 03
To describe the rates of pediatric antibiotic use across 6 countries on 3 continents.Cross-national analysis of 7 pediatric cohorts in 6 countries (Germany, Italy, South Korea, Norway, Spain, and the US) was performed for 2008-2012. Antibiotic dispensings were identified and grouped into subclasses. We calculated the rates of antimicrobial prescriptions per person-year specific to each age group, comparing the rates across different countries.A total of 74 744 302 person-years from all participating centers were included in this analysis. Infants in South Korea had the highest rate of antimicrobial consumption, with 3.41 prescribed courses per child-year during the first 2 years of life. This compares with 1.6 in Lazio, Italy; 1.4 in Pedianet, Italy; 1.5 in Spain; 1.1 in the US; 1.0 in Germany; and 0.5 courses per child-year in Norway. Of antimicrobial prescriptions written in Norway, 64.8% were for first-line penicillins, compared with 38.2% in Germany, 31.8% in the US, 27.7% in Spain, 25.1% in the Italian Pedianet population, 9.8% in South Korea, and 8% in the Italian Lazio population.We found substantial differences of up to 7.5-fold in pediatric antimicrobial use across several industrialized countries from Europe, Asia, and North America. These data reinforce the need to develop strategies to decrease the unnecessary use of antimicrobial agents.
Using Rich Data on Comorbidities in Case-Control Study Design with Electronic Health Record Data Improves Control of Confounding in the Detection of Adverse Drug Reactions.
D Backenroth, H Chase, C Friedman and Y Wei,
PloS one, 2016
Recent research has suggested that the case-control study design, unlike the self-controlled study design, performs poorly in controlling confounding in the detection of adverse drug reactions (ADRs) from administrative claims and electronic health record (EHR) data, resulting in biased estimates of the causal effects of drugs on health outcomes of interest (HOI) and inaccurate confidence intervals. Here we show that using rich data on comorbidities and automatic variable selection strategies for selecting confounders can better control confounding within a case-control study design and provide a more solid basis for inference regarding the causal effects of drugs on HOIs. Four HOIs are examined: acute kidney injury, acute liver injury, acute myocardial infarction and gastrointestinal ulcer hospitalization. For each of these HOIs we use a previously published reference set of positive and negative control drugs to evaluate the performance of our methods. Our methods have AUCs that are often substantially higher than the AUCs of a baseline method that only uses demographic characteristics for confounding control. Our methods also give confidence intervals for causal effect parameters that cover the expected no effect value substantially more often than this baseline method. The case-control study design, unlike the self-controlled study design, can be used in the fairly typical setting of EHR databases without longitudinal information on patients. With our variable selection method, these databases can be more effectively used for the detection of ADRs.
Update on Incidence of Herpes Zoster Among Children and Adolescents After Implementation of Varicella Vaccination, Antelope Valley, CA, 2000 to 2010.
R Civen, M Marin, J Zhang, A Abraham, R Harpaz, L Mascola and SR Bialek,
The Pediatric infectious disease journal, 2016 10
Changes in herpes zoster (HZ) epidemiology are expected with childhood varicella vaccination. We reported previously that during 2000 to 2006 HZ incidence decreased 55% in children <10 years of age, while among 10- to 19-year olds it increased by 63%. We update the analysis with 4 additional years of data.Population-based active surveillance was conducted for HZ in Antelope Valley, California. Structured telephone interviews and medical chart reviews collected data on demographics, varicella vaccinations, disease histories and clinical information. We calculated HZ incidence for 2007 to 2010 and assessed trends since 2000.Among children <10 years of age, HZ incidence continued the decreasing trend previously reported. During 2007 to 2010, the average incidence was 12.8 cases/100,000 children compared with 41.6 cases/100,000 children during 2000 to 2006, a 69% decline (P < 0.0001). For the 10- to 19-year olds, during 2007 to 2010 HZ incidence did not continue the increasing trend reported from 2000 to 2006; lower rates than in 2006 were observed in 3 of the 4 additional years evaluated. During 2007 to 2010 the average incidence was 78.2 cases/100,000 children compared with 68.0 cases/100,000 children during 2000 to 2006, a 13% increase (P = 0.123), with substantial fluctuation in annual rates throughout the 11 years of surveillance.During the mature varicella vaccination program, declines in HZ incidence among children <10 years of age continued through 2010. Among the 10- to 19-year olds, the increase reported through 2006 did not continue further and lower rates than in 2006 were observed through 2010. Widespread use of varicella vaccine could reduce HZ incidence among vaccinated populations. Ongoing monitoring of HZ incidence is needed to detect and understand changes in HZ epidemiology in the varicella vaccine era.