Creation of a Computable Phenotype in Childhood- Onset Arterial Ischemic Stroke (CAIS)
| dc.contributor | National Institutes of Health |
| dc.contributor.author | Dain, Aleksandra (Sarah) |
| dc.contributor.other | Children's Hospital of Philadelphia |
| dc.contributor.other | Nemours Children's Health |
| dc.date.accessioned | 2024-10-17T16:26:48Z |
| dc.description | The primary objective of this analysis is to develop a valid, automated method to identify children with CAIS using an EHR database, and to specifically ensure this method is valid in patients with SCD, who may present differently from others. We will use our validated CP to examine the epidemiology of stroke in SCD in a modern cohort, identify temporal trends in stroke rates, and examine risk factors for stroke in SCD. As TCD screening is the mainstay of stroke prevention in SCD, we will also use PEDSnet to perform a comparative assessment of TCD reporting strategies across sites and include TCD results as a variable in our descriptive analysis of stroke in SCD. #### Study Aims Aim 1: To derive and internally validate a computable phenotype to identify childhood arterial ischemic stroke in children and adolescents. - Hypothesis: This CP will have a minimum 85% positive predictive value (PPV) in single-center validation. Aim 2: To externally validate the CP in SCD-CAIS and all-cause CAIS. - Hypothesis: A single CP will have a minimum 85% PPV in both all-cause CAIS and SCD-CAIS. Aim 3. To perform a comparative assessment of TCD reporting strategies across all PEDSnet sites. - Hypothesis: Despite variation in TCD reporting strategies, we will be able to use automated methods to accurately ascertain TCD results among multiple sites. #### Study Design This is a retrospective cohort study to evaluate EHR markers of childhood arterial ischemic stroke in children and young adults with and without sickle cell disease, and to evaluate stroke risk factors for those with SCD treated in the current era. |
| dc.description.abstract | The primary aim of this study is to incorporate discrete data elements included in the PEDSnet database to create a valid computable phenotype for childhood arterial ischemic stroke (CAIS), using a rule-based approach. The phenotype should identify all types of childhood arterial ischemic stroke, an entity which is defined by multiple stroke subtypes and at-risk patient populations, especially in patients with sickle cell disease (SCD) as patients with SCD present to care frequently for pain and other disease complications, and presentation and evaluation of stroke in SCD may differ from all-cause stroke. |
| dc.identifier.uri | https://hdl.handle.net/20.500.14642/824 |
| dc.language.iso | en |
| dc.publisher | PEDSnet |
| dc.rights | a CC-BY 4.0 Attribution license. |
| dc.rights.uri | https://creativecommons.org/licenses/by-sa/4.0/ |
| dc.subject | PEDSnet Data Source |
| dc.subject | NIH P30 |
| dc.subject | Federally-Funded Research |
| dc.subject | Retrospective Study |
| dc.subject | Cohort Study |
| dc.subject.mesh | Ischemic Stroke |
| dc.subject.mesh | Stroke |
| dc.subject.mesh | Cardiovascular Diseases |
| dc.title | Creation of a Computable Phenotype in Childhood- Onset Arterial Ischemic Stroke (CAIS) |
| dspace.entity.type | Study |
| local.admin.note | Scientific Lead (may be first author): Aleksandra Sarah Dain Scientific Co-Lead (last author): Lauren Beslow PEDSnet Senior Scientist: Christopher Forrest Data Scientist/Biostatistician Lead (may be first author): Sahal Master Data Scientists/Biostatisticians: Project Manager: Elana DiCocco |
| local.code.github | https://github.com/PEDSnet/p3_CAIS |
| local.code.github | https://github.com/PEDSnet/p3_cais_phenotype |
| local.contributor.siteLead | Children's Hospital of Philadelphia |
| local.contributor.siteSponsor | Children's Hospital of Philadelphia |
| local.description.analytics | Study sample for Aims 1 and 2 is comprised of individuals aged 29 days to 210.9 years old with inpatient and/or ED encounters will be included in the creation and validation of the stroke computable phenotype. <br> **Inclusion Criteria:** - Individuals ages 29 days or older - Inpatient or emergency department (ED) encounter **Exclusion Criteria:** <br> None <br><br> For Aim 3, the study cohort is comprised of patients meeting the following criteria: <br> **Inclusion Criteria:** - SCD diagnosis, per Michalik et al. Requires: 1. A qualifying diagnosis code for SCD 2. 2 outpatient visits at least 30 days apart or 1 hospitalization in the EMR - Exclude if # diagnosis for sickle trait > qualifying sickle cell disease diagnoses - Two or more outpatient visits with a hematologist since January 2009 (is this included in our SCD definition?) **Exclusion Criteria:** - Age >= 18 on cohort entrance date - Age <2 on cohort exit date ##### Independent Variables The following variables will be assessed for inclusion in the Computable Phenotype: | Data Domain | Details | |------ | ------ | |Visits | Hospitalization <br>ED encounter <br>Specialist visit <br>Neurology <br>Hematology | |Procedures | Brain CT <br>Brain MRI <br>Blood transfusion <br>Thrombectomy <br>Angiogram | | Medications |Anticoagulants <br>Aspirin <br>Thrombolytics | |Labs | Hypercoagulability studies | { dqcheck.table } ##### Dependent Variables Primary dependent variables include CAIS, TCD vessel velocities, and TCD categorical results. |
| local.identifier.pedsnetid | 2024.DAIS.NIH.CHOP |
| project.endDate | Present |
| project.startDate | 2024 |
| relation.isOrgUnitOfStudy | cdb3cef9-ebdd-4ca8-9000-14573ba301bf |
| relation.isOrgUnitOfStudy | ff3cd76b-cf4c-44c1-84a6-368cfd0e9123 |
| relation.isOrgUnitOfStudy.latestForDiscovery | da841c56-bb55-405c-8b50-8f221ef62890 |
| relation.isPublicationOfStudy | 88dd6903-1cb0-4296-9ace-62eba94d7ff2 |
| relation.isPublicationOfStudy.latestForDiscovery | 88dd6903-1cb0-4296-9ace-62eba94d7ff2 |
| relation.isStudyOfStudy | 200be5c9-2075-4128-a778-a5517b650691 |
| relation.isStudyOfStudy.latestForDiscovery | 200be5c9-2075-4128-a778-a5517b650691 |