Background: While morbidity with surgical abortion is rare, later gestational age is a known risk factor for complications. Referral to a second healthcare provider delays abortion and this delay may result in significant increases in gestational age at the time of procedure. Referral may also increase the patient-incurred costs associated with obtaining an abortion.
No previous studies have examined the impact that referral to the hospital based setting has on patients seeking abortion care. Of particular interest to us is that women report referral for obesity and history of cesarean section, which for the most part do not require hospital-based care. To date no systematic evaluation has been conducted of obesity or history of cesarean section as factors associated with referral.
Statement of Purpose: This study was designed to compare sociodemographic and clinical factors between patients referred from outpatient abortion clinics to hospital-based care with those presenting directly for hospital-based care in Philadelphia, PA. We also sought to compare the associated care-delays and patient-level costs between these two groups. We hypothesized that obesity and history of cesarean delivery are associated with referral, and that referral is associated with care-delay and increased costs.
Methodology: In this cross-sectional study data were gathered through patient surveys and medical record review. Dependent variables included wait time, defined as date from first phone call to a provider to date of procedure, and patient-level costs for healthcare and associated expenses. A sample size of 85 participants per group provided us with 80% power to assess obesity, and 90% power to assess cesarean delivery as risk factors for referral, with a type I error of 5%. Multivariable logistic and generalized linear models were utilized to evaluate predictors of referral, wait time, and cost.   
Important Findings: Two-hundred-one participants were enrolled; 85 referred from outside abortion clinics, and 116 presenting directly to our hospital-based clinic. We found that the two referral groups differed significantly in every sociodemographic category except age. After controlling for differences in sociodemographics and health history, we found that morbid obesity was statistically significantly associated with referral from another facility (OR 7.51, p=0.008). Women referred from an outside facility had a median wait time of 28 days (IQR: 19-49), compared to 12 days (IQR:8-27, p=0.019) for non-referred patients.