Materials and Methods

The study population was identified retrospectively from a large national database containing clinical data collected prospectively from high-risk pregnant women receiving outpatient perinatal nursing services from Alere (Atlanta, GA; formerly Matria Healthcare). These services provide comprehensive outpatient services for conditions that place a pregnancy at risk for adverse outcomes. The population consisted of both commercially insured and Medicaid patients receiving care from obstetricians in either academic or community settings.

Each patient’s individual health care provider determined patient diagnoses, referral for outpatient services, timing, and method of delivery. Outpatient perinatal nursing services were provided in addition to care through the physician’s office. Consent for the use of each patient’s deidentified clinical data for research and reporting purposes is received during the enrollment process. The institutional review board of Central Baptist Hospital approved the analysis of this data. All data were collected using standardized definitions, policies, procedures, forms, and computer systems.

The outpatient hypertension surveillance program included an initial education session conducted by an experienced perinatal nurse in the patient’s home. Verbal instructions and written educational materials specific to the condition of gestational hypertension, as well as clinical signs and symptoms of a worsening condition, were provided. Each patient was instructed on self-care procedures, including use of equipment for collection of biometric clinical data. Patients were required to measure blood pressure using an automated device a minimum of twice daily. Urine protein testing was performed daily via dipstick. Maternal weight was measured daily. Fetal movement counts were performed daily and as needed. Patients received antenatal testing with nonstress test and/or biophysical profile evaluations 1 to 2 times per week at the discretion of their obstetric provider. Nursing staff contacted the patient per telephone at least 1 time per day to assess clinical data and question the patient about subjective maternal signs and symptoms (headache, visual changes, epigastric pain, and edema). Written and verbal reports were provided to the prescribing physician and case manager weekly or emergently as needed. As part of the program, 24/7 perinatal nursing support was available for patient questions or concerns.

Medical records of women with a singleton pregnancy enrolled in outpatient hypertension surveillance program between 1995 and 2007 were studied. Included for analysis were records from women with a diagnosis of mild gestational hypertension, no prior history of chronic hypertension, and documented pregnancy outcome at ≥34 weeks of gestation. Mild gestational hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg but <160 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg but <110 mmHg on at least 2 occasions 6 hours apart without the presence of proteinuria by urine dipstick analysis, without maternal symptoms of severe disease and with normal complete blood count and liver function tests.

Information regarding the patient’s medical and obstetric history, current pregnancy risk factors, diagnoses, medical and pharmacologic treatments and interventions, daily clinical assessments, and pregnancy outcome are collected from the patient and/or her healthcare provider and maintained in a relational database. Policies and procedures, forms, questionnaires, and a standardized electronic medical record are used at all patient care locations, contributing to homogeneity of data. Continuous quality improvement measures on a centralized basis and at all nursing call centers allows for monitoring of data integrity and nursing compliance.

After delivery, maternal and neonatal outcome data were collected by a perinatal nurse and recorded in the database. For purposes of the study, we categorized each patient into 1 of 3 “reason for delivery” groups; spontaneous preterm delivery, maternal/fetal indications, or elective/nonindicated delivery. We defined indicated delivery as the presence of 1 of the following: (1) spontaneous preterm labor, premature rupture of membranes (PROM) or vaginal bleeding, (2) maternal/fetal indications including development of preeclampsia, severe hypertension, hemolysis, elevated liver enzymes and low platelets (HELLP) syndrome or thrombocytopenia, diabetes, fetal growth restriction, or nonreassuring fetal testing. Elective delivery was defined as any delivery predicated by induced labor or scheduled cesarean without documentation of any of the above reasons for delivery. Maternal outcome variables studied included the cesarean section rate, gestational age at delivery, and mortality. Selected neonatal outcome variables included admission to the neonatal intensive care unit (NICU), neonatal respiratory complications (including respiratory distress syndrome [RDS] and need for ventilation such as mechanical ventilator or continuous positive airway pressure [CPAP]), presence of jaundice (need for phototherapy), and total neonatal hospital stay (NICU, intermediate, and regular nursery days). Decisions for admission to the NICU and management of neonatal complications were at the discretion of the attending neonatologists.

Outcomes were compared among those with nonindicated, elective delivery late-preterm vs 37 ^0/7 – 37 ^6/7 weeks and ≥38 ^0/7 weeks using Student t , Mann-Whitney U , or Pearson’s χ ² test statistics. Two-sided P values < .05 are shown and were adjusted for multiple comparisons (2 total groups vs control).