Profiling postpartum recovery following scheduled cesarean delivery with neuraxial anesthesia: A longitudinal cohort study

Abstract

BACKGROUND:

Childbirth can have a substantial impact on maternal health-related quality of life. Cesarean delivery is the most performed inpatient operation, yet little is known about normal postpartum recovery profiles. The primary aim of our study was to longitudinally evaluate global health visual analog scale (GHVAS; 0–100) scores up to 12 weeks following scheduled cesarean delivery and identify time to plateau of scores. Secondary aims were to evaluate different domains of postpartum recovery using validated patient-reported outcome measures (Obstetric Quality of Recovery score; ObsQoR-10 and 5-level 5-dimensional EuroQol questionnaire; EQ-5D).

METHODS:

After institutional review board approval, this single center, prospective longitudinal study enrolled healthy women scheduled for cesarean delivery. Women were excluded for gestational age < 32 weeks, neonatal demise, neonatal intensive care unit admission, inability to read or understand English, and if general anesthesia was used. Women completed baseline surveys prior to delivery and then at 24 and 48 hours after delivery. After hospital discharge, women completed surveys (including GHVAS, OBsQoR-10, EQ-5D, Edinburgh Postnatal Depression Scale, and activities of daily living) at 1 week, 3 weeks, 6 weeks, and 12 weeks postpartum. One-way repeated measures analysis of variance (ANOVA) was used to detect the difference in GHVAS and postpartum recovery outcomes with different follow up times points.

RESULTS:

We enrolled 66 parturients and 3 were withdrawn. Response rates were 95%, 84%, 83%, and 76% at 1, 3, 6, and 12 weeks, respectively. Mean ± SD GHVAS scores were 78±16 at baseline, 64±17 at 24 hours, 69±15 at 48 hours, 75±19 at 1 week, 88±11 at 3 weeks, 88±15 at 6 weeks, and 90±12 at 12 weeks postpartum (p<0.001). The global health VAS improved up until week 3 and then plateaued close to the maximum score between 3 weeks and 12 weeks postpartum. Mean ± SD ObsQoR10 scores were 75±15 at 24 hours, 85±10 at 48 hours, and 81±28 one week postpartum (p=0.003). The mean ± SD EQ-5D composite scores improved at 6 weeks (4.9±2.9) and 3 months (4.2±2.6) compared to baseline (6.5±1.8) with usual activities (p=0.001) and pain/discomfort (p<0.001) showing significant improvement over time. ObsQoR-10 score at 24 hours correlated with ObsQoR-10 scores at 48 hours (r=0.629, P <0.001) and 1 week (r=0.429, P <0.001) but did not correlate with EQ-5D scores at 6 weeks and 12 weeks.

CONCLUSIONS:

Our study demonstrates that GHVAS following scheduled CD plateaus at week 3. This data can be used to inform patients about anticipated trajectory of key postpartum recovery domains up to 12 weeks postpartum.

Introduction

Childbirth can have a substantial impact on maternal health-related quality of life. Recovery after childbirth and cesarean delivery (CD) is a complex process with multiple components including physical, psychological, and social domains.¹ Measuring recovery after delivery is difficult and requires an emphasis on patient-centered outcomes.^2–5 Traditionally, studies have assessed resolution of pain after cesarean delivery^6,7 but have not accounted for other domains essential to recovery such as patient experience, functional recovery, and quality of recovery after surgery.¹

The Obstetric Quality of Recovery score (ObsQoR-11 item, subsequently modified to a 10 item measure, ObsQoR-10) has been developed to measure recovery after childbirth, is validated following all delivery modes ^{8 9–11} and has been identified as the best measure of inpatient postpartum recovery.^4,5 The 5 dimensional EuroQol questionnaire (EQ-5D) is an extensively validated recovery instrument that measures 5 metrics including mobility, self-care, usual activities, pain or discomfort, and anxiety or depression, and is suited to assess longer term recovery after discharge. The global health visual analog scale (GHVAS) allows subjects to rate their overall postoperative recovery using a 100-mm visual analog scale and has been used as a global assessment of recovery in many surgical populations. Studies of recovery after childbirth have either focused only on the inpatient setting, or measured isolated time points postpartum. The longitudinal trajectory of postpartum recovery and how patient-reported outcome measure values change over time are not well elucidated.

The objective of the study was to determine trajectory of key postpartum recovery domains up to 12 weeks postpartum. The primary aim of our study was to longitudinally evaluate GHVAS scores up to 12 weeks following scheduled CD and identify time to plateau of scores. The ObsQoR-10 and EQ-5D both include a validated GHVAS score, which allowed both in- and outpatient recovery to be assessed utilizing the same scale. Secondary aims were to evaluate different domains of postpartum recovery using validated patient-reported outcome measures in the inpatient setting (ObsQoR-10) and after weeks/months after discharge (EQ-5D).

Methods:

The study was approved by the Mayo Clinic Institutional Review Board in Rochester, Minnesota (19–012077), and the protocol was registered prior to patient enrollment at ClinicalTrials.gov (NCT04462107, Principal Investigator: Emily Sharpe, Date of Registration: July 8, 2020). The study design was amended from validating ObsQoR-10 and EQ-5D to the current study design after registration but prior to patient enrollment. Adult patients were recruited from a single center (Mayo Clinic Hospital, Rochester, MN). Patients were approached for inclusion when study investigators were available for enrollment and survey follow up. Patients were enrolled and written informed consent was obtained prior to their CD. Women with singleton or multiple gestations scheduled for CD were included. Women were excluded from the study for gestational age less than 32 weeks, neonatal demise, neonatal intensive care unit (NICU) admission, inability to read or understand written English, and if general anesthesia was used. Demographic information, obstetric history, and fetal characteristics were collected by investigators not involved in the clinical care of the patient. Patients completed baseline surveys preoperatively, including EQ-5D, GHVAS (using a 10-cm line and measured in millimeters from 0 to 100; 0=the worst health imaginable and 100=the best health imaginable), and performance of activities of daily living (ADL) questions (assessing ADLs the day prior to delivery) via paper questionnaires. The ADL questionnaire was developed through an iterative process.³ (Supplemental Digital Content - 1)

Standard practice for enhanced recovery after CD included patient education regarding CD and pain control modalities available after delivery for scheduled CDs. Anesthetic care for CD followed routine practice. A multimodal analgesia regimen was utilized including spinal anesthesia or combined spinal-epidural anesthesia (intrathecal hyperbaric bupivacaine 12 to 13.5 mg, fentanyl 15 mcg, and morphine 150 mcg) intraoperatively. Postoperatively, all patients received scheduled 1,000 mg of acetaminophen orally every 6 hours and 15 mg of ketorolac intravenously every 6 hours for three doses, which was then replaced with 600 mg of ibuprofen orally. Oral oxycodone 5 mg was administered as needed for NRS pain scores 4–6 and 10 mg was administered for pain scores 7–10. Patients did not receive acetaminophen or any nonsteroidal anti-inflammatory drugs preoperatively or intraoperatively. Patients were treated for pain, pruritus, shivering, and nausea using institutional post-cesarean order sets. Additional post-delivery enhanced recovery elements included immediate initiation of a general diet, out of bed to chair greater than two hours and one walk on day of delivery, and removal of the urinary catheter within 12 hours of delivery. Additional information obtained from the electronic medical record included total opioid consumption, quantitative blood loss, requirement of blood transfusion, treatment of intraoperative and postoperative shivering, postoperative opioid consumption at 24 and 48 hours, time to first postoperative opioid administration, postoperative antiemetic requirement, length of stay, and non-maternal complications prolonging hospital discharge including NICU admission, neonatal IV antibiotics, social reasons, and hospital delays.

On the first (24 hour) and second (48 hour) postoperative days, patients reported GHVAS and completed ObsQoR-10 via paper questionnaires. At 1 week postpartum, patients received a link to an online survey via short message service (SMS) text message to complete questions related to pain scores, opioid use, ObsQoR-10, and GHVAS. At 3 weeks, 6 weeks, and 12 weeks postpartum, patients responded to surveys available via a link to an online survey from text message for EQ-5D-5L, GHVAS, and questions related to pain scores and opioid consumption. In addition to these questions, at 6 weeks, patients also completed Edinburgh Postnatal Depression Scale (EPDS), questions about ADL, and a modified recovery questionnaire. The modified recovery questionnaire asked about infant feeding, sleep, fatigue, bonding, parenting, sense of control, and social functioning.^12–16 Patients were contacted by additional text messages or email for reminders if they failed to complete the surveys. Data were collected by Research Electronic Data Capture (REDCap) database and Qualtrics (Qualtrics, Provo, UT, USA).

The primary aim of our study was to identify time to plateau of global health through longitudinal evaluation of GHVAS scores up to 12 weeks following scheduled CD. Secondary aims were to evaluate different domains of postpartum recovery using validated patient-reported outcome measures (ObsQoR-10 and EQ-5D), and explore correlations between inpatient (ObsQoR-10, length of stay) and outpatient (EQ-5D and EPDS) recovery metrics.

Statistical analysis

The sample size was calculated based on the primary outcome: the difference of GHVAS between baseline, 24 hours, 48 hours, 1 week, 3 weeks, 6 weeks, and 12 weeks after delivery. The median (IQR) of GHVAS has previously been reported to be 65 (50–80) on postpartum day 1.¹⁷ In order to detect an effect of partial eta squared of 0.03 (a small effect size) with 80% power in a one-way within-subject ANOVA (1 group, 7 measurements, alpha=0.05, non-sphericity correlation =0.5) with a null hypothesis that there is no difference in the population means of GHVAS at different time points, G*Power¹⁸ suggested we required 52 participants. We decided to recruit 63 participants to allow 20% loss of follow up.

Data were entered into Microsoft Excel 2013 (Microsoft Corporation, Redmond, WA). Data were analyzed using Stata Version 14.0 (StataCorp., College Station, TX). The Shapiro–Wilk test was used to assess for normal distribution of continuous variables. Mean (SD) or Median (IQR [range]) values were calculated for normally and non-normally distributed continuous variables, respectively. Categorical data were presented as count (percentage). One-way repeated measures analysis of variance (ANOVA) was used to detect the difference in postpartum recovery outcomes with different follow up times points (outcomes include: ObsQoR-10 and EQ-5D) and time was modeled as a categorical variable. Posthoc pairwise comparisons between two time points were made. Bonferroni correction was used to adjust for multiple comparison. Correlations between postpartum recovery domains were calculated using Pearson or Spearman correlation coefficients for normally and non-normally distributed data, respectively (0.4–0.7 considered moderate and 0.7–1.0 a strong correlation). Opioid medications used were converted into oral morphine mg equivalents by multiplying by a factor of 0.3 for IV fentanyl, 15 for IV hydromorphone, 1.5 for oral oxycodone, and 0.1 for oral tramadol.¹⁹ Statistical significance was set as a P value < 0.05 with 2-sided tests.

Results:

From February 2021 through July 2021, 127 patients were screened and 72 patients were approached about the study, six declined to participate, and 66 were enrolled. A total of 63 patients were included in the final analysis (Figure 1). Response rates were 95%, 84%, 83%, and 76% at 1, 3, 6, and 12 weeks postpartum, respectively. Demographic and obstetric characteristics are detailed in Table 1.

Figure 1:

Participant screening, recruitment and survey completion.

Table 1.

Demographics, obstetric and patient variables

	n = 63
Age (years)	33.2 ±3.6
Body Mass Index (kg/m2)	33.0 [28.6–38.1]
Race
Asian	2 (3.2%)
African American	3 (4.8%)
Hawaiian or other Pacific Island	1 (1.6%)
White	54 (85.7%)
Not reported	3 (4.8%)
Ethnicity
Not Hispanic or Latino	56 (88.9%)
Hispanic or Latino	4 (6.3%)
Chose not to disclose	3 (4.8%)
Education level
High school diploma	3 (4.8%)
Some college	6 (9.5%)
College graduate	30 (47.6%)
Graduate degree	24 (38.1%)
Insurance status
Private	57 (90.5%)
Medicaid	6 (9.5%)
ASA physical status classification
2	51 (81%)
3	11 (17.5%)
Missing	1 (1.5%)
Obstetric factors
Gestational age (weeks)	39 [38–39.2]
Parity
0	15 (23.8%)
1	30 (47.6%)
2	11 (17.5%)
3 or more	6 (11.1%)
Number prior cesarean deliveries
0	19 (30.2%)
1	44 (69.8%)
Indication for cesarean delivery
Repeat cesarean	44 (69.8%)
Breech	8 (12.7%)
Maternal co-morbidity (including pre-eclampsia and gestational diabetes)	1 (1.6%)
Multiple gestation	1 (1.6%)
Placenta previa	3 (4.8%)
Other	11 (17.5%)
Obstetric Comorbidities
Pre-eclampsia	1 (1.6%)
Gestational Diabetes Mellitus	10 (15.9%)
Diabetes Mellitus	2 (3.2%)
Gestational Hypertension	1 (1.6%)
Other	4 (6.4%)
Neonatal variables
Apgar at 1 min	8 [8,9]
Apgar at 5 min	9 [9,9]
Neonate weight	3450 ±603
Anesthesia technique
Spinal	52 (82.5%)
CSE	11 (17.5%)
QBL (mL)	601 [460, 1013]
Transfusion	0
Length of stay (delivery to discharge) (hours)	54.7 [52.3, 75.6]

Data presented as mean ± standard deviation, median [25^th, 75^th], or n (%)

NICU=neonatal intensive care unit; ASA=American Society of Anesthesiologists; CSE=combined spinal-epidural; QBL=quantitative blood loss.

Inpatient medications and recovery metrics are summarized in Supplemental Table 1. When evaluating non-maternal complications prolonging hospital discharge, there were 4 (6.4%) infants that required admission to the NICU and no documentation of neonatal IV antibiotics, social reasons for delay, or hospital factors for delay.

The mean ± SD GHVAS scores were 78±16 at baseline, 64±17 at 24 hours, 69±15 at 48 hours, 75±19 at 1 week, 88±11 at 3 weeks, 88±15 at 6 weeks, and 90±12 at 12 weeks postpartum (p<0.001). GHVAS significantly increased over time up to week 3 (48 hour scored 5 more than 24 hour after delivery (95% CI 1.1, 8.9), p=0.012; week 1 scored 11.2 more than 24 hour after delivery (95% CI 7.2, 15.2), p<0.001; week 3 after delivery scored 23.8 more than 24 hour after delivery (95% CI 19.7, 27.9), p<0.001), after which there was no significant increase in GHVAS postpartum scores (p=0.788 for week 6 vs week 3, p=0.261 for week 6 vs week 12) (Figure 2). The total score and individual components of the ObsQoR-10 at 24 hours, 48 hours, and 1 week postpartum are detailed in Supplemental Table 2. Scores increased from 24 to 48 hours (mean difference= 10.2 95% CI (4.3, 16.0); P <0.001), indicating better recovery. All components of ObsQoR-10 increased over time except patient comfort. The EQ-5D composite scores improved at 6 weeks and 3 months compared to baseline, with the individual components of usual activities and pain/discomfort showing significant improvement up to 12 weeks (Supplemental Table 3).

Figure 2: Global health VAS scores at baseline, 3 weeks, 6 weeks, and 12 weeks postpartum.

Box plot demonstrating global health VAS scores at various study time points. P<0.05 between baseline and 24 hours, 24 hours and 48 hours, 48 hours and 1 week, and 1 week and 3 weeks. VAS = visual analog scale

Resumption of ADL returned to pre-delivery baseline in most components, except driving, by 6 weeks postpartum (Table 2). After discharge, only 2% (1/52), 4% (2/51), and 0% (0/47) of patients reported taking opioids at 3 weeks, 6 weeks, and 12 weeks, respectively. EPDS and responses to a modified recovery questionnaire regarding infant feeding, sleep, fatigue, bonding, parenting, sense of control, and social functioning at 6 weeks postpartum are detailed in Supplemental Table 4.

Table 2:

Resumption of activities of daily living in outpatient postpartum period compared to pre-delivery baseline.

	Baseline	6 wk	P values
Current Activity Level	3.3±0.9	3.6±0.6	0.077
Current Exercise Level	2.0±0.9	1.8±0.7	0.083
Have you performed any of these activities in the last month?
Cooking	4.7±0.6	4.3±1.0	0.019
Attend events	2.6±1.3	2.7±1.4	0.667
Cleaning House	4.2±0.9	4.0±0.9	0.267
Dress yourself	5.0±0.2	5.0±0.1	1.000
Restroom by yourself	5±0	5±0	-
Manage 1 flight stairs	5.0±0.2	4.9±0.3	0.182
Exercise	3.7±1.2	2.9±1.3	0.008
Driving	4.7±0.8	3.9±1.1	<0.001
Shopping	3.1±0.8	2.9±1.2	0.195
Total	42.5±4.4	33.2±16.9	<0.001

Data presented as mean ± standard deviation; ADL=activities of daily living

P<0.005 is considered significant using Bonferroni correction.

Exploratory analysis of correlations between inpatient and outpatient recovery metrics are provided in Table 3. ObsQoR-10 score at 24 hours correlated with ObsQoR-10 scores at 48 hours (r=0.629, P <0.001) and 1 week (r=0.429, P <0.001) but did not correlate with EQ-5D scores at 6 weeks and 3 months (Table 3). In addition, there was a negative correlation between length of stay and ObsQoR-10 score at 24 (r=−0.351, p=0.005) and 48 hours (r=−0.348, p=0.005). There was no correlation between ObsQoR-10 scores at 24 hours, 48 hours, or 1 week and EQ-5D scores at 6 weeks and 12 weeks. EQ-5D scores at 6 weeks correlated with 12 week scores (r=0.419, P =0.001). Lower ObsQoR-10 scores at 24 hours (r=−0.311, p=0.026), 48 hours (r=−0.425, p=0.002), and 1 week (r=−0.435, p=0.001) (i.e. worse recovery) correlated with higher EPDS scores (i.e. depression screening score).

Table 3.

Correlation matrix for ObsQoR-10, length of stay, EQ-5D and EPDS

	ObsQoR-10 24 hour	ObsQoR-10 48 hour	ObsQoR-10 1 week	LOS	EQ-5D 6 week	EQ-5D 3 month	EPDS
ObsQoR-10 24 hour	-
ObsQoR-10 48 hour	0.629 (0.446 – 0.811) P<0.001	-
ObsQoR-10 1 week	0.429 (0.203 – 0.655) P=0.001	0.584 (0.397 – 0.770), P<0.001	-
LOS	−0.351 (−0.590 - −0.113) P=0.005	−0.348 (−0.601 - −0.096), P=0.005	−0.019 (−0.279 – 0.241), P=0.883	−
EQ-5D 6 week	−0.044 (−0.306 – 0.218), P=0.731	−0.034 (−0.309 – 0.240), P=0.789	−0.034 (−0.306 – 0.238), P=0.793	−0.113 (−0.345 – 0.120), P=0.380	−
EQ-5D 3 month	−0.133 (−0.384 – 0.118), P=0.299	0.112 (−0.153 – 0.378), P=0.407	0.082 (−0.193 – 0.357), p=0.523	−0.040 (−0.292 – 0.213), P=0.757	0.419 (0.146 – 0.693), P=0.001	−
EPDS	−0.311 (−0.545 - −0.077), P=0.026	−0.425 (−0.618 - −0.233), P=0.002	−0.435 (−0.665 - −0.204), P=0.001	0.102 (−0.178 – 0.381), p=0.478	0.467 (0.235 – 0.698), P=0.001	0.447 (0.220 – 0.675), p=0.001	−

Values are Spearman’s rank correlation coefficient (95% Confidence Interval). Correlations were calculated using Spearman correlation coefficients for non-normally distributed data (0.4–0.7 considered moderate and 0.7–1.0 considered a strong correlation). ObsQoR-10 = Obstetric Quality of Recovery score; LOS = length of stay; EQ-5D = 5-level 5-dimensional EuroQol questionnaire; EPDS = Edinburgh Postnatal Depression Scale; LOS=length of stay.

DISCUSSION:

Our study demonstrates that when evaluated longitudinally, GHVAS after scheduled CD significantly improves up to 3 weeks and then plateaus close to the maximum score, but ADLs have not returned to predelivery baseline by 6 weeks postpartum. Inpatient recovery improves in the first 2 days postpartum and inpatient ObsQoR scores correlate moderately with 1-week ObsQoR-10 score and correlate weakly with 6 week EPDS score and hospital length of stay but not EQ-5D scores.

Traditionally, studies have evaluated the trajectory of pain resolution following delivery. Komatsu et al. found the median (IQR) days for pain-free functional recovery was 14 (7 to 24) days after vaginal delivery and 21 (14 to 27) days after CD.⁷ Booth et al. identified 6 clusters of recovery patterns in over 500 patients after CD and pain declined initially in a log(time) fashion for 2 to 4 weeks and then a change-point occurred with most clusters showing resolution of pain, yet persistent pain occurred in 9% of patients at 2 months.⁶ Our study is the first to demonstrate the majority of global recovery, as assessed by GHVAS, occurs by 3 weeks postpartum following scheduled CD.

To assess other components of recovery beyond pain, studies have evaluated recovery of activity. Recovery of physical activity after delivery has been measured both by self-report⁷ and using objective measures such as accelerometry.²⁰ Komatsu et al. observed functional recovery of pre-pregnancy activity level by self-report to be 57 days in 95% of patients after CD.⁷ Sharpe et al.²⁰ report an increase in daytime step count over the first 2 months postpartum in a curvilinear pattern with considerable variability between individuals and an inverse correlation between pain and step count. Massouh et al. evaluated the correlation between ambulation (using accelerometry) and quality of recovery (using QoR-15) after CD and found a moderate correlation between ambulation and QoR-15 score; however, patients were only evaluated in the first 24 hours postpartum.²¹ We did not objectively measure physical activity recovery in this study, but EQ-5D and activity questionnaires queried resumption of mobility and activities and found improvement over baseline.

The ObsQoR-10 is considered the best measure of inpatient postpartum recovery following vaginal and operative deliveries^4,5 and is part of a core outcome set evaluating enhanced recovery after CD.²² Most studies measure recovery using the ObsQoR in the first 24 to 48 hours postpartum but Mazda et al. queried postpartum patients daily for 7 days and found recovery plateaued at 3 days after vaginal delivery and 4 days after scheduled CD.²³ We found improvement in ObsQoR-10 between 24 and 48 hours postpartum but similar scores at 1 week. A recent study demonstrated inpatient ObsQoR-10 scores correlate with 6 week EPDS scores.²⁴ Our results confirm this finding in the US setting. Studies utilizing targeted interventions in patients with lower inpatient ObsQoR scores are needed to determine if trajectories of postpartum recovery can be improved.

Few studies have evaluated quality of recovery beyond 1 week after hospital discharge. A study evaluating quality of recovery at 1, 3, and 6 weeks after delivery using EQ-5D and the Short-Form 36 (SF36) questionnaires found patients had recovery to baseline values by 3 weeks and 6 weeks after vaginal delivery and scheduled CD, respectively.²⁵ Petrou et al. compared recovery using EQ-5D score by modes of delivery at 12 months postpartum and found CD was associated with poorer long-term quality of recovery compared to spontaneous vaginal delivery.²⁶ In our study, the EQ-5D scores represented improved recovery at 6 weeks and 12 weeks over pre-delivery baseline with usual activities and pain/discomfort showing improvement over time. The EQ-5D does not evaluate all postpartum recovery domains including breastfeeding, infant health, motherhood experience, fatigue, psychosocial support, sleep, and sexual function.¹ Many patients are not seen by an obstetrician after 6 weeks postpartum, and most working women return to work by 12 weeks. Resumption of ADL had not returned to baseline by 6 weeks postpartum and were not assessed at the 12-week timepoint in our study. Further, there was no correlation between ObsQoR-10 and EQ-5D scores in our study but making comparisons is inherently flawed as they assess different domains of postpartum recovery.

Our study confirms that patients after scheduled CD have better response rates utilizing text messaging compared to email, phone calls, or paper data entry.⁶ We had a comparable compliance rate (76%−95%) to that described by Booth et al. (82%).⁶ This is higher than the 48% success rate reported for daily phone calls after delivery.⁷

Strengths of this study include the high response rate after hospital discharge. The study has several limitations. We studied relatively healthy patients, who were mostly non-Hispanic white, well-educated, and from a single hospital, limiting generalizability. In addition, our results are not generalizable to patients having cesarean delivery under general anesthesia. Patients who received general anesthesia were excluded as they may have higher postoperative pain scores and later mobilization, possibly impacting quality of recovery.²⁷ Pre-pregnancy values were not measured and so baseline values were taken on the day of delivery prior to their scheduled cesarean. The ObsQoR-10 has been validated to be used postpartum but the EQ-5D GHVAS scores have not specifically been developed or designed for use in postpartum patients; however, these outcome measures have been extensively validated in many healthcare settings and languages. Further, we were unable to control how the GHVAS scale was displayed on each patient’s device. Additionally, the EQ-5D mobility question was not included for the week 3 survey and so we were unable to calculate a total score at 3 weeks. Lastly, the study was performed in 2021 during the COVID-19 pandemic, and it is unknown if this impacted any specific domains of postpartum recovery.

Additional research is needed to determine if inpatient recovery measures can predict long-term recovery and outcomes and to characterize recovery using multidimensional measures capable of assessing all relevant postpartum recovery domains. Further work is needed to determine generalizability in socioeconomically, racially, and ethnically diverse populations. Future studies are needed to elucidate when different recovery domains plateau following scheduled CD and compare recovery following different delivery modalities. Furthermore, studies are needed to identify the most common domains responsible for worse recovery at different time points and the relationship between recovery domains at each time point.

Our study demonstrates GHVAS following scheduled CD plateaus at week 3, ADLs have not returned to baseline by week 6, and inpatient recovery metrics correlate with 1-week ObsQoR score. The inpatient ObsQoR scores weakly correlate with hospital length of stay and 6-week EPDS score but not EQ-5D scores. This data can be used to inform patients about anticipated trajectory of key postpartum recovery domains up to 12 weeks postpartum.

Supplementary Material

Supplemental Digital Content 1

Supplementary Digital Content 1 : Activities of Daily Living Questionnaire

Supplemental Table 1

Supplemental Table 1: Inpatient medications and recovery metrics

Supplemental Table 2

Supplemental Table 2. ObsQoR-10 breakdown at each time point (24 hours, 48 hours and 1 week postpartum) with longitudinal comparisons for each patient using paired data (24h vs 48h vs 1 week)

Supplemental Table 3

Supplemental Table 3. EQ5D longitudinal comparison at baseline, 3 weeks, 6 weeks, and 3 months postpartum.

Supplemental Table 4

Supplemental Table 4. Postpartum recovery questions at 6 weeks

Key Points Summary.

Question:

What is the profile of recovery longitudinally and how do global health visual analog scale (GHVAS) scores change for patients up to 12 weeks following scheduled cesarean delivery?

Findings:

The mean ± SD GHVAS scores were 78±16 at baseline, 64±17 at 24 hours, 69±15 at 48 hours, 75±19 at 1 week, 88±11 at 3 weeks, 88±15 at 6 weeks, and 90±12 at 12 weeks postpartum; inpatient ObsQoR scores correlate with ObsQoR at 1 week but not EQQ-5D at 6 weeks and 12 weeks.

Meaning:

The GHVAS improved up until week 3 and then plateaued up to week 12 postpartum; this data can be used to counsel patients about anticipated trajectory of key postpartum recovery domains up to 12 weeks postpartum.