Potential Confounders and EffectModifiersDiabetes, considered a likely effect modifier,was categorized into 4 groups: type 1diabetes, type 2 diabetes, gestational diabetes,and no diabetes. We further investigated thecombination of gestational and type 2 diabetes,because diabetes diagnosed duringpregnancy is often thought to be type 2 diabetesthat is identified through prenatal testing.30Definitions for diabetes were based onICD-9-CM codes as follows: type 1 diabetes and gestational diabetes (648.8).We obtained information about each patient’sage and type of medical insurancefrom hospital discharge records. Age was categorizedinto 5 groups: 15 to 17, 18 to 19,20 to 34, 35 to 44, and 45 to 54. Medicaidstatus was defined as being insured by Medicaidor enrolled in a Medicaid health maintenanceorganization. Because having no healthinsurance usually indicates both low incomeand delayed application to Medicaid for pregnantwomen,31 we combined women whosedischarge records indicated self-pay as themethod of payment (<5%) with womenwhose services were covered by Medicaid forour analyses.Finally, we stratified all analyses in thestudy by the geographic region of residenceas indicated by the county of residence onthe hospital discharge record. NYC includedthe city’s 5 counties: Bronx, Kings (Brooklyn),New York (Manhattan), Queens, and Richmond(Staten Island). NYS included all othercounties grouped together.
การแปล กรุณารอสักครู่..
