Table 1.
ICEES feature variables: name, description, and binning strategy
| Feature Variable a | Description and Binning Strategy |
|---|---|
| CohortID | Cohort ID# & input variables used to define cohort |
| PatientID | Randomly assigned patient identifier |
| StudyPeriod | Years 2010, 2011, 2012, 2013, 2015 |
| AgeStudyStart | Age at “study” start date: calculated from birth date and binned as 0…2, 3…17, 18…34, 35…50, 51…69, 70–89 years b |
| Sex | Male, Female, Unknown |
| Race | Caucasian, African American, Asian, Native Hawaiian/Pacific Islander, American/Alaskan Native, Other |
| Ethnicity | Hispanic (1=Yes, 0=No) |
| ObesityICD | ICD code c for obesity anytime over “study” period (1=Yes, 0=No) |
| ObesityBMI | BMI ≥ 30 anytime over “study” period (1=Yes, 0=No) |
| Diagnosis: AsthmaDx, CroupDx, ReactiveAirwayDx, CoughDx, PneumoniaDx | One or more diagnoses of select respiratory disorders (defined by high-level ICD categories) over “study” period; 1=Yes, 0= No for each diagnosis) |
| EstResidentialDensity | US Census Bureau ACS 2012-2016 estimated total population [block group], binned according to US Census Bureau definitions (1=rural [0, 2500), 2=urban cluster [2500, 50000), 3=urbanized area [50000, inf)) |
| EstResidentialDensity25Plus | US Census Bureau ACS 2012-2016 estimated total population aged 25 years or older [block group], binned as quintiles (1, 2, 3, 4, 5) |
| EstProbabilityNonHispWhite | US Census Bureau ACS 2012-2016 estimated proportion of persons who are non-Hispanic white [block group], binned as quartiles (1, 2, 3, 4) |
| EstProbabilityHouseholdNonHispWhite | US Census Bureau ACS 2012-2016 estimated proportion of households that are non-Hispanic white [block group], binned as quartiles (1, 2, 3, 4) |
| EstProbabilityHighSchoolMaxEducation | US Census Bureau ACS 2012-2016 estimated proportion persons aged 25 years or older with a HS diploma or less at their highest level of schooling [block group], binned as quartiles (1, 2, 3, 4) |
| EstProbabilityNoAuto | US Census Bureau ACS 2012-2016 estimated proportion of households without an automobile [block group], binned as quartiles (1, 2, 3, 4) |
| EstProbabilityNoHealthIns | US Census Bureau ACS 2012-2016 estimated proportion of persons without health insurance [block group], binned as quartiles (1, 2, 3, 4) |
| EstProbabilityESL | US Census Bureau ACS 2012-2016 estimated proportion of persons 5 years or older who sometimes speak a language other than English at home [block group], binned as quartiles (1, 2, 3, 4) |
| EstHouseholdIncome | US Census Bureau ACS 2012-2016 estimated median household income [block group], binned as quintiles (1, 2, 3, 4, 5) |
| MajorRoadwayHighwayExposure | US Census TIGERline distance in meters from household to nearest major road/highway (1 = 0-49, 2 = 50-99, 3 = 100-199, 4 = 200-299, 5 = 300-499, 6 = >=500 meters) |
| RoadwayDistanceExposure | US DOT distance in meters from household to nearest roadway (1 = 0-49, 2 = 50-99, 3 = 100-199, 4 = 200-299, 5 = 300-499, 6 = >=500 meters) |
| RoadwayType | UNC DOT roadway classification (eg, major highway) |
| RoadwayAADT | US DOT annual average daily traffic estimate |
| RoadwaySpeedLimit | US DOT roadway speed limit |
| RoadwayLanes | UNC DOT roadway number of lanes |
| AvgDailyPM2.5Exposure | UNC IE average estimated average daily PM2.5 exposure over “study” period, binned by values (1, 2, 3, 4, 5) |
| MaxDailyPM2.5Exposure | UNC IE average estimated maximum daily PM2.5 exposure over “study” period, binned by values (1, 2, 3, 4, 5) |
| AvgDailyOzoneExposure | UNC IE average estimated average daily ozone exposure over “study” period, binned by values (1, 2, 3, 4, 5) |
| MaxDailyOzoneExposure | UNC IE average estimated maximum daily ozone exposure over “study” period, binned by values (1, 2, 3, 4, 5) |
| Medications Prescribed or Administered: Prednisone, Fluticasone, Mometasone, Budesonide, Beclomethasone, Ciclesonide, Flunisolide, Albuterol, Metaproterenol, Diphenhydramine, Fexofenadine, Cetirizine, Ipratropium, Salmeterol, Arformoterol, Formoterol, Indacaterol, Theophylline, Omalizumab, Mepolizumab | One or more prescriptions/administrations of medication over “study” period (1=Yes, 0=No for each medication) |
| TotalEDInpatientVisits | Total number ED/inpatient visits for respiratory issue(s) (defined by same ICD codes used to pull patients with asthma-like conditions) over “study” period (0, 1, 2, 3, …) |
ACS: American Community Survey; BMI: body mass index; DOT: Department of Transportation; ED: emergency department; ICD: International Classification of Diseases; ICEES: Integrated Clinical and Environmental Exposures Service; IE: Institute for the Environment; PM25: particulate matter ≤2.5 µm in diameter; UNC: University of North Carolina.
The feature variables listed in the table are those for the patient-level tables, which include data on each patient for each year of available data (ie, data on individual patients are represented as rows in the table). Similar feature variables are available for the visit-level tables, although the variables are sometimes treated differently. For example, PM2.5 exposures are expressed in relation to the 24-hour and 2-week period before visits, not in relation to the 1-year “study” period, as was done for the patient-level tables. Additional feature variables (eg, laboratory measures) are available for select years. Further information can be accessed via the ICEES OpenAPI.
Maximum age is 89 years, per Health Insurance Portability and Accountability Act regulations.
The ICD codes for obesity and for the selection of patients with asthma-like conditions can be found in the Supplementary Appendix.