The research activity was approved by the University of Pittsburgh Institutional Review Board, and the requirement for written informed consent was waived. Medical records of all 2,494 women who delivered at Magee-Womens Hospital over the period February 1, 2015 to May 1, 2015 were each assessed for eligibility and had data abstracted by two investigators (LF and GL). Data that were abstracted from the medical record included: estimated gestational age, gravidity, parity, body mass index (BMI) based on height and weight measured on admission to the labor and delivery unit, mode of delivery (spontaneous vaginal, instrumental vaginal including forceps- or vacuum-assisted deliveries, or cesarean), use of epidural labor analgesia, pain scores during labor delivery, duration of labor (for spontaneous labor, defined as the documented time of rupture of membranes, or the time of admission to the labor and delivery unit, until the time of delivery; for induction of labor, defined as the start time of first induction medication administration, or the time of insertion of transcervical balloon, until the time of delivery), maternal co-morbid disease documented in the clinical notes or coded in the diagnostic problem list by ICD-9 or SNOMED Clinical Terms (e.g. hypertensive disorders, antepartum anemia, chronic pain, history of miscarriage, known fetal anomalies), psychiatric disorders (e.g. anxiety/depression, psychological trauma, bipolar disorder, other psychiatric diagnoses), and perineal injuries at delivery. Abuse was defined as a history of substance, partner, sexual, or childhood abuse; trauma was defined as a history of accidental, birth, or other trauma. Chronic pain was defined as a history of fibromyalgia, pseudotumor cerebri, inflammatory bowel or pelvic disease requiring medication, chronic back pain, or juvenile or rheumatoid arthritis. Data reliability were assessed by a two-person verification process; after one investigator completed data abstraction from the records, the second investigator reviewed the records as well as the abstracted data for consistency. Any discrepancies were resolved by consensus. Data were recorded in an Excel spreadsheet (Microsoft Inc., USA) and patient identifiers were removed prior to analysis.
Women who received epidural analgesia for labor pain, who had pain assessed during labor both before and at least once during implementation of labor epidural analgesia by 0–10 numeric rating scores (NRS), and who had depression risk assessed by the Edinburgh Postnatal Depression Scale (EPDS) at their six-week postpartum visit, were included in the final analysis. The EPDS is self-completed, 10-item scale developed specifically for women in the perinatal period. It has been shown to be an effective means of identifying patients at risk for perinatal depression.7 (link),8 (link) Women who did not have the primary outcome, EPDS score, recorded at the six-week postpartum visit were excluded from the primary analysis (Figure 1 ).
Women typically requested epidural labor analgesia at their discretion and were interviewed by the anesthesia service at the time of request. Over the investigation period, the institutional rate of utilization of labor epidural analgesia was approximately 90%. Initiation of labor epidural analgesia typically occurred by a loss of resistance to saline technique, followed by delivery of a test dose of lidocaine 1.5% with epinephrine 1:200,000 (3 mL), and finally by a bolus of epidural bupivacaine 0.083% with fentanyl 2 mcg/mL (8 mL) and epidural fentanyl 100 mcg in divided doses. Maintenance of analgesia typically occurred by patient-controlled epidural analgesia with bupivacaine 0.083% with fentanyl 2 mcg/mL at 8mL/hour continuous infusion, 8 mL bolus every 8 minutes by patient demand, with a 24mL/hour maximum. During labor, patients were asked to rate their labor pain intensity using a 0–10 NRS by their bedside nurse, where 0 is no intensity at all and 10 is the most intensity that can be imagined. Pain ratings are generally expected to be recorded in the medical record every one to three hours by the bedside nurse.
Percent improvement in pain (PIP) was used as the primary predictor.9 –11 (link) In sum, PIP is the percent change in pain over time. It is defined as the difference between baseline pain score and the average change in pain per unit of time, and is expressed as a percentage: PIP = ([baseline pain score −average intrapartum pain score]/baseline pain score) × 100. For example, a woman in labor with a baseline NRS of 7, and post analgesia NRS of 0 at 1 hour, 0 at 2 hours, and 4 at 3 hours will have an average post-analgesia pain of 1.33 per hour. In this case, PIP = ( [7 – 1.33] / 7 )*100 = 81%. With PIP, it is also possible that pain worsens after an analgesic intervention. An example of such a case would be a woman with a baseline NRS of 5, and post-analgesia NRS of 8 at 1 hour, 8 at 2 hours and 9 and 3 hours, resulting in an average post-analgesia pain of 8.33 per hour. In this case, PIP = ([5 – 8.33] / 5)*100 = −67%. Baseline pain NRS was defined as the pain score recorded immediately prior to the recorded time of labor epidural analgesia initiation. Women who did not have the primary predictor, PIP, calculable due to missing pain scores were excluded from the primary analysis, but were included in a sensitivity analysis after imputations were computed as described below.
Women who received epidural analgesia for labor pain, who had pain assessed during labor both before and at least once during implementation of labor epidural analgesia by 0–10 numeric rating scores (NRS), and who had depression risk assessed by the Edinburgh Postnatal Depression Scale (EPDS) at their six-week postpartum visit, were included in the final analysis. The EPDS is self-completed, 10-item scale developed specifically for women in the perinatal period. It has been shown to be an effective means of identifying patients at risk for perinatal depression.7 (link),8 (link) Women who did not have the primary outcome, EPDS score, recorded at the six-week postpartum visit were excluded from the primary analysis (
Women typically requested epidural labor analgesia at their discretion and were interviewed by the anesthesia service at the time of request. Over the investigation period, the institutional rate of utilization of labor epidural analgesia was approximately 90%. Initiation of labor epidural analgesia typically occurred by a loss of resistance to saline technique, followed by delivery of a test dose of lidocaine 1.5% with epinephrine 1:200,000 (3 mL), and finally by a bolus of epidural bupivacaine 0.083% with fentanyl 2 mcg/mL (8 mL) and epidural fentanyl 100 mcg in divided doses. Maintenance of analgesia typically occurred by patient-controlled epidural analgesia with bupivacaine 0.083% with fentanyl 2 mcg/mL at 8mL/hour continuous infusion, 8 mL bolus every 8 minutes by patient demand, with a 24mL/hour maximum. During labor, patients were asked to rate their labor pain intensity using a 0–10 NRS by their bedside nurse, where 0 is no intensity at all and 10 is the most intensity that can be imagined. Pain ratings are generally expected to be recorded in the medical record every one to three hours by the bedside nurse.
Percent improvement in pain (PIP) was used as the primary predictor.9 –11 (link) In sum, PIP is the percent change in pain over time. It is defined as the difference between baseline pain score and the average change in pain per unit of time, and is expressed as a percentage: PIP = ([baseline pain score −average intrapartum pain score]/baseline pain score) × 100. For example, a woman in labor with a baseline NRS of 7, and post analgesia NRS of 0 at 1 hour, 0 at 2 hours, and 4 at 3 hours will have an average post-analgesia pain of 1.33 per hour. In this case, PIP = ( [7 – 1.33] / 7 )*100 = 81%. With PIP, it is also possible that pain worsens after an analgesic intervention. An example of such a case would be a woman with a baseline NRS of 5, and post-analgesia NRS of 8 at 1 hour, 8 at 2 hours and 9 and 3 hours, resulting in an average post-analgesia pain of 8.33 per hour. In this case, PIP = ([5 – 8.33] / 5)*100 = −67%. Baseline pain NRS was defined as the pain score recorded immediately prior to the recorded time of labor epidural analgesia initiation. Women who did not have the primary predictor, PIP, calculable due to missing pain scores were excluded from the primary analysis, but were included in a sensitivity analysis after imputations were computed as described below.
