Abstract
Introduction
Pregnant individuals with cardiometabolic conditions—including diabetes, hypertensive disorders, and obesity—are at elevated risk for postpartum complications. Timely postpartum care is critical for mitigating postpartum risks, yet postpartum visit attendance remains suboptimal, and few evidence‐based interventions exist, especially among those receiving care in safety‐net health systems. We aimed to evaluate whether a patient‐centered postpartum care planning intervention increases comprehensive postpartum visit attendance among obstetric patients with at least one cardiometabolic risk factor.
Methods
We conducted a single‐center, parallel group randomized control trial at a metropolitan safety‐net health system, enrolling pregnant individuals who were 20–34 weeks gestation, English or Spanish‐speaking, and had a diagnosis of diabetes mellitus, chronic hypertension, gestational diabetes, gestational hypertension or preeclampsia, and/or pre‐pregnancy obesity. Participants were randomized 1:1 to receive standard obstetric care (control) or standard obstetric care plus prenatal education and co‐creation of a postpartum care plan (intervention). The primary outcome was attendance at a comprehensive postpartum visit between 4 and 12 weeks postpartum. We used generalized linear models to estimate risk ratios (RRs) and 95% confidence intervals (CIs) for the intervention using an intent‐to‐treat approach, stratified by parity.
Results
We randomized 159 participants to the intervention and 161 to the control arms (N = 320). Overall, 52.8% and 55.3% of participants in the intervention and control arms, respectively, attended a comprehensive postpartum visit (p = 0.66). When stratified by parity, no differences were seen among primiparous participants (61.5% vs. 50.0%; RR, 1.23; 95% CI, 0.82–1.85) or multiparous participants (48.6% vs. 56.6%; RR, 0.86; 95% CI, 0.67–1.10). For multiparous participants, among those receiving one or no prenatal study visits with a research coordinator, only 29.6% attended a comprehensive postpartum visit, whereas 72.4% of those with ≥3 visits did so (p = 0.001).
Conclusion
The patient‐centered postpartum care planning intervention did not significantly improve postpartum visit attendance. However, attendance rates were higher among those with greater prenatal intervention exposure, suggesting that intensity and continuity of patient engagement may improve postpartum care utilization.
Trial registration: Date of registration: June 20, 2022; date of initial participant enrollment: August 8, 2022; clinical trial identification number: NCT05430815; URL of the registration site: https://clinicaltrials.gov/study/NCT05430815
Keywords: gestational diabetes, obesity, patient‐centered care, postpartum period, pregnancy, pregnancy complications, pregnancy‐induced hypertension
1. INTRODUCTION
The 12 months following delivery (the postpartum period) constitute a critical window of physical and mental health transitions and increased need for social and emotional support [1, 2]. Nearly two‐thirds of US maternal deaths occur between 1 and 365 days postpartum [3], and approximately 1%–2% of postpartum patients are readmitted within 42 days, with recent increases in these rates [4, 5]. Rates of severe maternal morbidity have also risen [6], and approximately 27% of events occur during the postpartum year [7]. The burden of postpartum morbidity and mortality disproportionately affects non‐Hispanic Black and American Indian or Alaska Native individuals and those with lower educational attainment, covered by Medicaid, or lacking insurance at delivery [4, 8, 9].
In 2018, the American College of Obstetricians and Gynecologists (ACOG) published recommendations reframing postpartum care as an ongoing, patient‐centered planning process that is initiated during prenatal care, reinforces the importance and value of postpartum care, and culminates in a comprehensive visit and transition to well‐woman care [1]. However, postpartum visit attendance in the United States remains suboptimal, at an estimated 72% nationally, with the lowest rates among those who identify as non‐Hispanic Black or Hispanic, have lower socioeconomic status, and are publicly insured [10, 11].
Postpartum care is especially crucial for individuals with cardiometabolic conditions, including obesity, diabetes, and hypertensive disorders, given their elevated risk for postpartum complications [1]. Despite recommendations for targeted follow‐up (blood pressure and/or glucose monitoring, counseling on cardiovascular risk, and linkage to primary or specialty care) [1], postpartum care attendance remains low, particularly for populations facing systemic barriers such as fragmented care systems, transportation barriers, limited care coordination, and unmet social, physical, and mental health needs [12, 13, 14].
Although ACOG recommends antepartum education, care coordination, and postpartum care planning, rigorous evaluation of these strategies is limited [15, 16]. To address this gap, we conducted a parallel‐group randomized controlled trial of standard obstetric care versus standard care plus a patient‐centered postpartum care planning intervention on postpartum visit attendance among pregnant individuals with at least one cardiometabolic risk factor [17].
2. METHODS
2.1. Study setting
This parallel group randomized controlled trial was conducted at Grady Memorial Hospital, a large public hospital in Atlanta, Georgia, with approximately 3000 deliveries annually. The protocol was approved by Emory University's Institutional Review Board and Grady's Research Oversight Committee and was registered on clinicaltrials.gov (NCT05430815). An independent data and safety monitoring board provided additional oversight.
2.2. Eligibility
Pregnant individuals between 20 and 34 weeks’ gestation who attended ≥1 prenatal care visit, intended to deliver at our institution, were fluent in English or Spanish, and had a diagnosis of diabetes mellitus, chronic hypertension, gestational diabetes, a hypertensive disorder of pregnancy (gestational hypertension or preeclampsia), and/or prepregnancy obesity (body mass index [BMI] of 30 or greater), documented in the electronic medical record (EMR), were eligible. Individuals who were incarcerated or unable to consent were excluded.
2.3. Intervention and procedures
Consenting participants were randomized 1:1 (unblinded) to the intervention or control arm using simple randomization generated in R (“randomizeR” package) [18] and implemented in Research Electronic Data Capture (REDCap). Personnel who enrolled participants did not have access to the allocation sequence.
Details of the intervention have been published and are summarized in Figure S1 [17]. Briefly, the goal of the intervention was to develop a postpartum care plan that reflected individual needs and preferences, with clinically relevant needs documented in the EMR and accessible to the patient via an EMR portal (MyChart) and in printed after‐visit summaries. The goal of the plan was to enhance patient engagement in care, encourage attendance at the postpartum visit, and facilitate the transition to primary care. During routine prenatal care visits, intervention participants attended study visits with a trained research coordinator (licensed practical nurse) who engaged patients in individualized education and postpartum planning around 20 topics related to health and social needs (including pregnancy complications, immunizations, reproductive life planning, mental health, postpartum support, returning to work, healthy weight, substance use, food and housing security, infant feeding, heart disease prevention, chronic disease management, and transitioning to primary care; see Table S1 for a full list).
Study visits occurred before or after scheduled prenatal visits and were spread throughout prenatal care to minimize burden. All participants received the same educational content; however, the number of topics covered at each study visit varied based on timing of entry into the study and the number of prenatal care visits that were attended. Supplementing the sessions were language‐concordant printed materials maintained in a binder provided to participants and a web‐based resource library accessible via QR code. During the visits, the coordinator reviewed each of the educational modules and provided navigation and care coordination to address unmet medical and social care needs, including housing instability, intimate partner violence, and transportation barriers. The postpartum care plan, which reflected evidence‐based recommendations and complemented the educational content, was documented in the EMR using a Smartform and was available to patients via MyChart and a printed visit summary. Intervention participants could complete up to eight study visits and four supplemental visits. Phone or text message reminders were sent before each scheduled study visit.
The coordinator also met with intervention participants during their delivery hospitalization to review and finalize the care plan and followed up by phone 1‐week postpartum to assess needs and link with resources. A small number of intervention participants (n = 3) enrolled later in pregnancy and did not have time to attend any prenatal study visits; these individuals only received a single postpartum planning visit where education was provided and a postpartum care plan was developed and documented. Attending physicians and residents were informed of the postpartum care plan through email reminders every 6 weeks to coincide with resident rotations and received in‐person reminders, encouraging them to review the care plan during postpartum care. Control participants did not attend study visits or have a separate care plan documented in the EMR. For all participants, the postpartum visit was scheduled by the nurses at the hospital prior to delivery discharge at around 6 weeks postpartum.
At enrollment, all participants completed a sociodemographic survey. At 12‐weeks postpartum, all participants were asked to complete a web‐based survey assessing contraceptive use, self‐rated physical and mental health status, depression symptoms (using the Edinburgh Postnatal Depression Scale [EPDS]), use of primary care, and perceived risk for future cardiovascular disease and adverse pregnancy outcomes. Up to 10 contact attempts were made to complete the survey.
Intervention participants were provided incentives at enrollment ($25), after each study visit ($10), and after completing the surveys ($25). Control participants received incentives at enrollment ($25) and after survey completion ($25).
2.4. Outcomes
The primary outcome was attendance at a comprehensive postpartum visit between 4 and 12 weeks after delivery, defined as a preventive care visit with an obstetric care provider (e.g., obstetrician‐gynecologist, certified nurse midwife, or primary care provider with additional training relevant to postpartum care) assessing multiple aspects of physical and mental health. Candidate visits were identified via EMR review that included visits within our health system or other health systems using shared EMR or via requested records and were adjudicated by two blinded independent obstetricians to determine consistency with the outcome definition; disagreements were resolved by a third physician reviewer.
Secondary outcomes included any type of health care visit between 0 and 3 weeks and 4 and 12 weeks postpartum, number of readmissions or emergency department/urgent care visits through 12 weeks postpartum (ascertained from medical records, including primary diagnosis), self‐reported contraception use, EPDS score, mental and physical health, and perceived future health risks for cardiovascular disease, adverse pregnancy outcomes, stroke, and diabetes (all ascertained by survey). Reasons for readmission and emergency department/urgent care visits were abstracted from the EMR and classified by a study physician. Readmissions due to severe maternal morbidity (SMM) were identified using published criteria [18]. EPDS score was dichotomized at ≥13, which is a reliable indicator of probable depressive symptoms [19]. Self‐reported measures of mental and physical health were reported using scores of 1 (excellent health) through 6 (poor health). Participants were asked to rank their risk of future complications into three categories of high, average, or low; responses were categorized as low versus average or high.
2.5. Statistical analysis
Based on a priori power calculations using historical data from our institution [12], we assumed a 55% postpartum visit attendance rate in controls and 10% loss to follow‐up. We estimated that 320 participants were needed to achieve 80% power to detect at a risk ratio (RR) of ≥1.5 for the primary outcome of visit attendance. As national estimates suggest that 72% of women attend their postpartum visit [10] and other studies of postpartum care navigation resulted in a threefold increase in odds of visit attendance [20], we felt that an RR of 1.5 was feasible in our population, especially given our low initial rates.
Baseline demographic and clinical characteristics were compared for the intervention versus control groups using χ2 tests. We used log‐binomial regression to estimate RRs and 95% confidence intervals (CIs) for all binomial outcomes and linear regression to calculate parameter estimates and 95% CIs for continuous outcomes. Due to skewness from excess zeros, we used zero‐inflated Poisson models to estimate effects on readmissions and emergency department/urgent care visits. We identified effect measure modification by parity and stratified all outcomes by primiparous and multiparous status. Analyses followed an intent‐to‐treat approach and were conducted using SAS version 9.4, with alpha = 0.05 as the threshold for statistical significance. There were no missing data for the primary outcome. Missing data for the secondary outcomes of self‐reported health status and perceived risks were excluded using a complete case analysis.
3. RESULTS
From August 2022 to October 2023, 2804 pregnant individuals receiving prenatal care at our institution were screened (Figure 1). The primary exclusion was not meeting study criteria (e.g., 20 and 34 weeks’ gestation, attended ≥1 prenatal care visit, intended to deliver at our institution, fluent in English or Spanish, and had a diagnosed cardiometabolic condition, n = 2248); then inability to contact the patient (e.g., appointment no‐shows, n = 168) or declined participation (n = 68). A total of 320 patients were randomized to standard care (n = 161) or standard care plus the postpartum care planning intervention (n = 159). Two intervention participants withdrew (due to lack of time and interest) but were included in analyses. One withdrew immediately after enrollment and the other after attending three study visits.
FIGURE 1.
Study flow diagram. Note: all participants were included in the analysis of the primary outcome even if they withdrew from the study.
Baseline characteristics were balanced between groups, except for parity and prenatal care visits (Table 1). Approximately one‐third (32.7%) of intervention participants were primiparous versus 19.9% of controls (p = 0.09). Intervention participants were less likely to initiate care in the first trimester (62.9% vs. 76.8%, p = 0.009) and had fewer prenatal care visits than controls (13.3 vs. 16.1, p = 0.005). Most participants identified as non‐Hispanic Black (82.6%), were enrolled in Medicaid or Medicare (78.4%), and had a diagnosis of prepregnancy obesity (46.6%) and/or hypertension (43.8%). Among intervention participants, 38.4% had one study visit, 30.2% had two visits, and 31.4% had three or more visits.
TABLE 1.
Demographic and clinical characteristics of the study population at enrollment, enrolled August 2022–October 2023.
| Intervention (n = 159) | Control (n = 161) | ||
|---|---|---|---|
| N (%) | N (%) | p value | |
| Age at enrollment (years) | |||
| 18–24 | 42 (26.4) | 37 (23.0) | 0.45 |
| 25–34 | 81 (50.9) | 78 (48.4) | |
| 35–45 | 36 (22.6) | 46 (28.6) | |
| Race and ethnicity | 0.90 | ||
| Non‐Hispanic Black | 133 (83.6) | 128 (81.5) | |
| Non‐Hispanic white | 3 (1.9) | 3 (1.9) | |
| American Indian | 0 | 1 (0.6) | |
| Hispanic (any race) | 23 (14.5) | 25 (15.9) | |
| Unknown | 0 | 4 | |
| Insurance at delivery | 0.28 | ||
| Medicaid/Medicare | 119 (74.8) | 132 (82.0) | |
| Private | 35 (22.0) | 27 (16.8) | |
| Self‐pay | 1 (0.6) | 1 (0.6) | |
| Unknown | 4 (2.5) | 1 (0.6) | |
| Gestational age at enrollment | 0.80 | ||
| 20–27 weeks | 98 (61.6) | 97 (60.2) | |
| 28–34 weeks | 61 (38.4) | 64 (39.7) | |
| Diagnosis | |||
| Chronic hypertension | 70 (44.0) | 70 (43.5) | 0.92 |
| Diabetes mellitus | 18 (11.3) | 21 (13.0) | 0.64 |
| Prepregnancy obesity | 79 (49.7) | 70 (43.5) | 0.27 |
| Gestational diabetes | 18 (11.3) | 24 (14.9) | 0.34 |
| Hypertensive disorder of pregnancy | 6 (3.8) | 8 (5.0) | 0.60 |
| Parity following index birth | |||
| 1 | 52 (32.7) | 32 (19.9) | 0.09 |
| 2 or more | 107 (67.3) | 129 (80.1) | |
| Prenatal care visits | |||
| Number of visits (mean, SD) | 13.3 (8.0) | 16.1 (9.9) | 0.005 |
| Initiated care in first trimester | 100 (62.9) | 123 (76.8) | 0.009 |
| Number of prenatal study visits a | |||
| ≤1 | 61 (38.4) | ||
| 2 | 48 (30.2) | ||
| 3 or more | 50 (31.4) |
Abbreviation: SD, standard deviation.
Only includes study visits prior to delivery. Some participants had only one study visit that occurred during the delivery hospitalization.
Overall, 54.1% (n = 173) attended a comprehensive postpartum care visit 4–12 weeks after delivery (primary outcome), with differences by parity. Among primiparous participants, 61.5% (n = 32) of intervention versus 50.0% (n = 16) of controls attended their visit (RR, 1.23; 95% CI, 0.82–1.85) (Table 2). Among multiparous participants, there was no difference in postpartum visit attendance for intervention (48.6%, n = 52) versus controls (56.6%, n = 75) (RR, 0.86; 95% CI, 0.67–1.10). Similar patterns were observed for any postpartum visit during 4–12 weeks’ postpartum. The rate of visits between 0 and 3 weeks was similar for both intervention primiparous and multiparous participants (RR, 0.89; 95% CI, 0.65–1.23 and RR, 0.87; 95% CI, 0.68–1.10, respectively).
TABLE 2.
Postpartum visit attendance, readmissions, and emergency department or urgent care visits through 12 weeks postpartum, stratified by parity at the index birth.
| Primiparous following index birth | Multiparous following index birth | |||||
|---|---|---|---|---|---|---|
| Intervention (n = 52) | Control (n = 32) | Intervention (n = 107) | Control (n = 129) | |||
| N (%) | N (%) | RR (95% CI) | N (%) | N (%) | RR (95% CI) | |
| Attended 4–12 week comprehensive postpartum visit (primary outcome) | ||||||
| Yes | 32 (61.5) | 16 (50.0) | 1.23 (0.82–1.85) | 52 (48.6) | 73 (56.6) | 0.86 (0.67–1.10) |
| No | 20 (38.5) | 16 (50.0) | 1.00 | 55 (51.4) | 56 (43.4) | 1.00 |
| Attended any 4–12 week postpartum visit | ||||||
| Yes | 38 (73.1) | 20 (62.5) | 1.17 (0.85–1.60) | 71 (66.4) | 96 (74.4) | 0.89 (0.75–1.05) |
| No | 14 (26.9) | 12 (37.5) | 1.00 | 36 (33.6) | 33 (25.6) | 1.00 |
| Attended any visit 0–3 weeks | ||||||
| Yes | 32 (61.5) | 22 (68.8) | 0.89 (0.65–1.23) | 54 (50.5) | 75 (58.1) | 0.87 (0.68–1.10) |
| No | 20 (38.5) | 10 (31.3) | 1.00 | 53 (49.5) | 54 (41.9) | 1.00 |
| Number of readmissions | ||||||
| 0 | 49 (94.2) | 31 (96.9) | 2.56 (0.23–28.9) a | 104 (97.2) | 121 (93.8) | 0.40 (0.10–1.55) a |
| 1 | 2 (3.8) | 1 (3.1) | 3 (2.8) | 7 (5.4) | ||
| >1 | 1 (1.9) | 0 | 0 | 1 (0.8) | ||
| Number of emergency department or urgent care visits | ||||||
| 0 | 37 (71.1) | 25 (78.1) | 2.16 (0.86–5.43) a | 66 (80.4) | 109 (84.5) | 1.47 (0.80–2.70) a |
| 1 | 10 (19.2) | 7 (21.9) | 16 (14.9) | 15 (11.6) | ||
| >1 | 5 (9.6) | 0 | 5 (4.7) | 5 (3.9) | ||
Abbreviations: CI, confidence interval; RR, risk ratio; SD, standard deviation.
RR estimated using zero‐inflated Poisson regression, which is used to analyze count data with excess of zero values.
Primiparous participants in the intervention arm had a total of four readmissions compared with one in the control arm (aRR, 2.56; 95% CI, 0.23–28.9). Among multiparous participants, there were three and nine readmissions for the intervention and controls, respectively (RR, 0.40; 95% CI, 0.10–1.55). There were a total of 52 emergency department/urgent care visits among primiparous participants receiving the intervention versus 32 in primiparous controls (RR, 2.16; 95% CI, 0.86–5.43). Multiparous intervention participants had 107 emergency department/urgent care visits compared with 129 in the multiparous controls (RR, 1.47; 95% CI, 0.80–2.70). There were five SMM readmissions, with two (3.8%) occurring in the primiparous intervention group and three (2.3%) in the multiparous controls. The reasons for emergency department/urgent care visits were similar for both arms (Table S2), with the most frequent diagnoses being acute health conditions unrelated to pregnancy, hypertension‐related issues, and wound infections and/or dehiscence.
When the rates of comprehensive postpartum visit attendance were stratified by the number of study visits attended, attendance increased with increasing number of study visits for multiparous participants: from 29.6% with one or fewer visits to 72.4% with ≥3 study visits (p = 0.001) (Figure 2). Among primiparous participants, the highest attendance (66.7%) was also among those with ≥3 study visits, though patterns were less consistent.
FIGURE 2.
Rates of comprehensive postpartum visit attendance among 159 participants in the intervention arm, stratified by the number of prenatal study visits attended and parity. p = 0.001 for the distribution of postpartum visit attendance by number of study visits for multiparous participants only.
Rates of 12‐week survey completion were high in both groups (88.7% intervention vs. 87.6% control) (Figure 1). Contraceptive use was reported by 87.4% (n = 236/270) with no differences by group or parity (Table 3). Among primiparous participants, EPDS scores ≥13 (indicating probable depression) were reported in 4.9% of those receiving the intervention versus 14.8% of the controls (RR, 0.33; 95% CI, 0.06–1.67). No significant group difference was seen among multiparous participants (17.1% vs. 18.3%; aRR 0.93; 95% CI, 0.49–1.77). About 30.7% of participants overall reported attending a checkup with their primary care provider, with no difference by group. Likewise, self‐reported mental and physical health and perceived future health risks were similar between groups regardless of parity. Among primiparous participants, the intervention group was slightly more likely to perceive themselves at low risk for serious future pregnancy complications and diabetes, although differences were not significant.
TABLE 3.
Self‐reported health and perceived risk outcomes at 12 weeks postpartum (among 282 participants that completed the survey).
| Primiparous following index birth | Multiparous following index birth | |||||
|---|---|---|---|---|---|---|
| Intervention (n = 47) | Control (n = 29) | Intervention (n = 94) | Control (n = 112) | |||
| N (%) | N (%) | RR (95% CI) | N (%) | N (%) | RR (95% CI) | |
| Using contraception | ||||||
| Yes | 42 (93.3) | 25 (86.2) | 1.08 (0.92–1.28) | 77 (85.6) | 92 (86.8) | 0.99 (0.88–1.10) |
| No | 3 (6.7) | 4 (13.8) | 1.00 | 13 (14.4) | 14 (13.2) | 1.00 |
| Refused/missing | 2 | 0 | 4 | 6 | ||
| EPDS score | ||||||
| ≥13 | 2 (4.9) | 4 (14.8) | 0.33 (0.06–1.67) | 14 (17.1) | 17 (18.3) | 0.93 (0.49–1.77) |
| <13 | 39 (95.1) | 23 (85.2) | 1.00 | 65 (82.9) | 76 (81.7) | 1.00 |
| Refused/missing | 6 | 2 | 12 | 19 | ||
| Attended checkup with primary care provider | ||||||
| Yes | 15 (33.3) | 10 (37.0) | 0.90 (0.47–1.71) | 25 (30.5) | 26 (27.7) | 1.10 (0.69–1.75) |
| No | 30 (66.7) | 17 (63.0) | 1.00 | 57 (69.5) | 68 (72.3) | 1.00 |
| Refused/missing | 2 | 2 | 12 | 18 | ||
| Mental health since delivery a | ||||||
| Mean (SD) | 2.6 (1.5) | 2.5 (1.5) | 0.09 (−0.61 to 0.80) b | 2.6 (1.5) | 2.5 (1.5) | 0.12 (−0.31 to 0.55) b |
| Don't know/refused/missing | 0 | 0 | 3 | 1 | ||
| Physical health since delivery a | ||||||
| Mean (SD) | 2.5 (1.5) | 2.5 (1.2) | 0.05 (−0.63 to 0.73) b | 2.4 (1.2) | 2.3 (1.3) | 0.12 (−0.25 to 0.48) b |
| Don't know/refused/missing | 0 | 2 | 4 | 5 | ||
| Perceived risk of having a serious complication in a future pregnancy | ||||||
| Low | 18 (51.4) | 9 (37.5) | 1.37 (0.75 to 2.52) | 17 (25.8) | 25 (29.1) | 0.89 (0.52–1.50) |
| High or average | 17 (48.6) | 15 (62.5) | 1.00 | 49 (74.2) | 61 (70.9) | 1.00 |
| Don't know/refused/missing | 12 | 5 | 28 | 26 | ||
| Perceived risk of my baby having serious complications in a future pregnancy | ||||||
| Low | 18 (52.9) | 8 (36.4) | 1.45 (0.77–2.75) | 25 (35.7) | 33 (38.8) | 0.92 (0.61–1.39) |
| High or average | 16 (47.1) | 14 (63.6) | 1.00 | 45 (35.7) | 52 (61.2) | 1.00 |
| Don't know/refused/missing | 13 | 7 | 24 | 27 | ||
| Perceived risk of having a stroke in the next 15 years | ||||||
| Low | 28 (82.3) | 16 (72.7) | 1.13 (0.84–1.53) | 27 (57.4) | 42 (62.7) | 0.92 (0.67–1.25) |
| High or average | 6 (17.6) | 6 (27.3) | 1.00 | 20 (42.5) | 25 (37.3) | 1.00 |
| Don't know/refused/missing | 13 | 7 | 47 | 45 | ||
| Perceived risk of being diagnosed with diabetes in the next 15 years | ||||||
| Low | 21 (60.0) | 8 (44.4) | 1.35 (0.75–2.42) | 31 (55.4) | 39 (52.7) | 1.05 (0.76–1.44) |
| High or average | 14 (40.0) | 10 (55.6) | 1.00 | 25 (44.6) | 35 (47.3) | 1.00 |
| Don't know/refused/missing | 12 | 11 | 38 | 38 | ||
Abbreviations: CI, confidence interval; EPDS, Edinburgh Postnatal Depression Scale; RR, risk ratio; SD, standard deviation.
Responses scored on a Likert‐type scale where 1 = excellent and 6 = poor.
Parameter estimates were derived from a linear regression model.
4. DISCUSSION
This single‐center, parallel group randomized control trial found that a patient‐centered postpartum care planning intervention did not significantly increase the primary outcome of postpartum visit attendance among the full study population or when stratified by parity. However, we found that postpartum visit attendance increased significantly with the number of prenatal study visits, rising from 29.6% to 72.4% among multiparous participants with ≥3 study visits. These findings highlight the importance of sustained prenatal care engagement and care continuity in motivating postpartum care utilization, particularly among multiparous women who may be less likely to attend postpartum visits [11, 21].
Our findings differ from an observational postpartum patient navigation study in an urban, predominantly Medicaid‐enrolled population that found a threefold increase in postpartum visit attendance among those who received navigation [20]. That study enrolled participants during the delivery hospitalization and provided concentrated support after delivery, whereas our intervention was primarily prenatal. Conversely, our results align with a randomized trial of intensive nurse home visitation for Medicaid‐eligible individuals that found no effect on postpartum contraceptive use or family planning visit attendance in the year after delivery [22].
Our study adds to the scant evidence on the impact of postpartum care planning and suggests that while structured, individualized care during pregnancy is important, continued support throughout the postpartum period is likely essential for sustaining long‐term care engagement. Qualitative studies identify lack of preparation for the postpartum period as a barrier to care engagement, particularly for individuals with chronic conditions, pregnancy complications, or social needs [23, 24, 25]. One study of a comprehensive postpartum clinic for high‐risk pregnancies found 84% attended a postpartum visit and 61% had multiple visits, with success largely attributed to community health workers linking patients to social, mental health, and primary care services [26].
Emerging data suggest that postpartum navigation may reduce severe maternal morbidity and improve mental health outcomes [27, 28]. In our study, we found some evidence of higher rates of postpartum readmissions and emergency department/urgent care visits among primiparous participants who received the intervention versus the controls, possibly due to earlier recognition and care‐seeking for complications. However, these estimates were imprecise and should be interpreted with caution. Likewise, there was a trend toward lower rates of postpartum depression (EPDS score ≥13) among primiparous intervention participants compared with controls, perhaps reflecting improved prenatal linkage to mental health care and identification of a maternal support team as part of the postpartum care plan.
While we did not find a significant effect on postpartum visit rates, our findings show the feasibility of integrating proactive, patient‐centered care planning into routine obstetric workflows. Flexible models adapting to parity, risk factors, and resource access are critical to improving postpartum outcomes. The association between increased prenatal contact and postpartum care uptake highlights the importance of consistent communication, potentially delivered via trusted care coordinators or support staff. Further, tailored prenatal education may increase appropriate care‐seeking. Embedding these efforts into routine workflows may address known barriers to postpartum engagement, including poor care coordination, lack of anticipatory guidance, and unmet social or behavioral health needs.
Future research should evaluate targeted planning interventions and ways to promote continuity and follow up with planning visits. Larger studies powered for subgroup analyses by parity, race/ethnicity, insurance, and language are needed to guide targeted strategies. Additional studies of multimodal interventions integrating telehealth, web‐based education and support, and home visitation into postpartum care planning are needed, particularly for populations facing systemic barriers to care [15, 16]. Intervention design must also address provider and health system factors such as bias and discrimination, racial and language concordance, health literacy, and logistical barriers to scheduling and attending appointments [11, 15, 23]. Researchers should use implementation science frameworks and assess the feasibility, acceptability, and sustainability of patient‐centered planning models across diverse settings.
Strengths of this study include its randomized design, focus on high‐risk obstetric population, and use of electronic medical records to assess primary outcomes. The use of a patient‐centered, pregnancy‐specific intervention and parity‐stratified analysis provides important insights into differential responses.
Limitations include the single‐center design and inclusion of only English‐ or Spanish‐speaking participants, which may limit generalizability. The individualized nature of intervention makes it difficult to isolate the most effective components. Increased prenatal engagement was associated with greater postpartum care uptake, though this likely reflects participant self‐selection or greater baseline engagement. The trial was unblinded; therefore, group assignment may have influenced the outcome. There were differences in the distribution of parity and prenatal care use for the intervention versus control participants, which may be a consequence of the simple randomization approach [29]. The intervention concluded at 1‐week postpartum, potentially limiting longer‐term impacts. While survey response rates were balanced across arms (∼88%), missing data may introduce bias. However, the use of EMR data for primary outcomes reduces reliance on self‐report. Participants in the intervention arm received greater incentives, which could have contributed to differences in postpartum care engagement. Finally, although we observed effect modification by parity, the study was not powered to detect differences within strata, and estimates should be interpreted with caution.
Despite awareness of the need to improve postpartum care models, rigorous data to support evidence‐based approaches are limited. Our study contributes to this critical knowledge gap and suggests that a patient‐centered comprehensive care planning system is feasible and has the potential to improve outcomes but may require different approaches for primiparous versus multiparous patients. Moving forward, pairing tailored postpartum care planning during pregnancy with care navigation in the year following delivery may be a promising strategy for reducing care fragmentation and promoting long‐term engagement and health.
AUTHOR CONTRIBUTIONS
Sheree L. Boulet: Conceptualization; data curation; investigation; formal analysis; methodology; funding acquisition; writing—original draft preparation; project administration; software; supervision. Kaitliyn K. Stanhope: Conceptualization; data curation; investigation; methodology; funding acquisition; writing—review and editing; project administration; software. Taylore King: Investigation; writing—review and editing; validation. Franklyn Geary: Conceptualization; investigation; writing—review and editing. Anna Newton‐Levinson: Conceptualization; investigation; writing—review and editing. Marisela Lozada: Investigation; writing—review and editing; data curation. Taé Stallworth: Writing—review and editing; project administration; data curation; investigation; supervision. Danielle Vuncannon: Investigation; validation; writing—review and editing. Victoria L. Green: Conceptualization; investigation; writing—review and editing. Gabriela Juarez: Writing—review and editing; investigation; data curation. Lauren Furst: Investigation; writing—review and editing; validation. Denise J. Jamieson: Conceptualization; investigation; funding acquisition; writing—review and editing; project administration; supervision; resources. Anne L. Dunlop: Conceptualization; investigation; funding acquisition; writing—review and editing; project administration; supervision; data curation; resources.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
ETHICS STATEMENT
This project was approved by the Emory University Institutional Review Board and the Grady Research Oversight Committee.
Supporting information
Supporting Information
Supporting Information
ACKNOWLEDGEMENTS
We would like to thank Nakai Brown, Lauren Costley, N. Esther Osbourne, and Ugochukwu Ukponu for their contributions to this project. We would like to acknowledge the support of the Grady Health System, Atlanta, Georgia, in conducting this research. This study is funded by the National Institute of Minority Health and Health Disparities (R01MD016031).
This study was presented as an oral presentation at the 2024 APPAM Fall Research Conference, November 21–23, 2024.
DATA AVAILABILITY STATEMENT
The data from this trial will be made available to researchers upon reasonable request to the corresponding author. Investigators will be required to submit a data analysis request form that requires elucidation of the specific data requested along with a statement that the investigator must provide: (1) a commitment to using the data only for research purposes and not to identify any individual participant, (2) a commitment to securing the data using appropriate computer technology, and (3) a commitment to destroying or returning the data after analyses are completed. Fully deidentified data will be provided after approval by the corresponding author and other members of the study team, as appropriate.
REFERENCES
- 1. 2018. “ACOG Committee Opinion No. 736: Optimizing Postpartum Care.” Obstetrics and Gynecology 131(5): e140–e50. 10.1097/AOG.0000000000002633. [DOI] [PubMed] [Google Scholar]
- 2. Tully, K. P. , Stuebe A. M., and Verbiest S. B.. 2017. “The Fourth Trimester: A Critical Transition Period With Unmet Maternal Health Needs.” American Journal of Obstetrics and Gynecology 217: 37–41. 10.1016/j.ajog.2017.03.032. [DOI] [PubMed] [Google Scholar]
- 3. Centers for Disease Control and Prevention . 2024. Pregnancy‐Related Deaths: Data From Maternal Mortality Review Committees in 36 US States, 2017–2019. Accessed August 19, 2025. https://www.cdc.gov/maternal‐mortality/php/data‐research/mmrc‐2017‐2019.html.
- 4. Tucker, C. M. , Ma C., Mujahid M. S., Butwick A. J., Girsen A. I., Gibbs R. S., and Carmichael S. L.. 2024. “Trends in Racial/Ethnic Disparities in Postpartum Hospital Readmissions in California From 1997 to 2018.” AJOG Global Reports 4: 100331. 10.1016/j.xagr.2024.100331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Clapp, M. A. , Little S. E., Zheng J., and Robinson J. N.. 2016. “A Multi‐State Analysis of Postpartum Readmissions in the United States.” American Journal of Obstetrics and Gynecology 215: 113.e1–e10. 10.1016/j.ajog.2016.01.174. [DOI] [PubMed] [Google Scholar]
- 6. Fink, D. A. , Kilday D., Cao Z., Larson K., Smith A., Lipkin C., Perigard R., et al. 2023. “Trends in Maternal Mortality and Severe Maternal Morbidity During Delivery‐Related Hospitalizations in the United States, 2008 to 2021.” JAMA Network Open 6: e2317641. 10.1001/jamanetworkopen.2023.17641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Boghossian, N. S. , Greenberg L. T., Buzas J. S., Rogowski J., Lorch S. A., Passarella M., Saade G. R., and Phibbs C. S.. 2024. “Severe Maternal Morbidity From Pregnancy Through 1 Year Postpartum.” American Journal of Obstetrics & Gynecology MFM 6: 101385. 10.1016/j.ajogmf.2024.101385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Fleszar, L. G. , Bryant A. S., Johnson C. O., Blacker B. F., Aravkin A., Baumann M., Dwyer‐Lindgren L., et al. 2023. “Trends in State‐Level Maternal Mortality by Racial and Ethnic Group in the United States.” JAMA 330: 52–61. 10.1001/jama.2023.9043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wang, E. , Glazer K. B., Howell E. A., and Janevic T. M.. 2020. “Social Determinants of Pregnancy‐Related Mortality and Morbidity in the United States: A Systematic Review.” Obstetrics and Gynecology 135: 896–915. 10.1097/AOG.0000000000003762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Attanasio, L. B. , Ranchoff B. L., Cooper M. I., and Geissler K. H.. 2022. “Postpartum Visit Attendance in the United States: A Systematic Review.” Women's Health Issues 32: 369–75. 10.1016/j.whi.2022.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wouk, K. , Morgan I., Johnson J., Tucker C., Carlson R., Berry D. C., and Stuebe A. M.. 2021. “A Systematic Review of Patient‐, Provider‐, and Health System‐Level Predictors of Postpartum Health Care Use by People of Color and Low‐Income and/or Uninsured Populations in the United States.” Journal of Women's Health 30: 1127–59. 10.1089/jwh.2020.8738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Campbell, A. , Stanhope K. K., Platner M., Joseph N. T., Jamieson D. J., and Boulet S. L.. 2022. “Demographic and Clinical Predictors of Postpartum Blood Pressure Screening Attendance.” Journal of Women's Health 31: 347–55. 10.1089/jwh.2021.0161. [DOI] [PubMed] [Google Scholar]
- 13. D'Amico, R. , Dalmacy D., Akinduro J. A., Hyer M., Thung S., Mao S., Fareed N., and Bose‐Brill S.. 2023. “Patterns of Postpartum Primary Care Follow‐Up and Diabetes‐Related Care After Diagnosis of Gestational Diabetes.” JAMA Network Open 6: e2254765. 10.1001/jamanetworkopen.2022.54765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Interrante, J. D. , Admon L. K., Carroll C., Henning‐Smith C., Chastain P., and Kozhimannil K. B.. 2022. “Association of Health Insurance, Geography, and Race and Ethnicity with Disparities in Receipt of Recommended Postpartum Care in the US.” JAMA Health Forum 3: e223292. 10.1001/jamahealthforum.2022.3292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Whyler, N. C. A. , Krishnaswamy S., Price S., and Giles M. L.. 2024. “Strategies to Improve Postpartum Engagement in Healthcare After High‐Risk Conditions Diagnosed in Pregnancy: A Narrative Review.” Archives of Gynecology and Obstetrics 310: 69–82. 10.1007/s00404-024-07562-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Phillips, S. E. K. , Celi A. C., Wehbe A., Kaduthodil J., and Zera C. A.. 2023. “Mobilizing the Fourth Trimester to Improve Population Health: Interventions for Postpartum Transitions of Care.” American Journal of Obstetrics and Gynecology 229: 33–38. 10.1016/j.ajog.2022.12.309. [DOI] [PubMed] [Google Scholar]
- 17. Stanhope, K. K. , Stallworth T., Forrest A. D., Vuncannon D., Juarez G., Boulet S. L., Geary F., et al. 2024. “Planning for the Forgotten Fourth Trimester of Pregnancy: A Parallel Group Randomized Control Trial to Test a Postpartum Planning Intervention vs. Standard Prenatal Care.” Contemporary Clinical Trials 143: 107586. 10.1016/j.cct.2024.107586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Centers for Disease Control and Prevention . Identifying Severe Maternal Morbidity (SMM). Accessed August 19, 2025. https://www.cdc.gov/maternal‐infant‐health/php/severe‐maternal‐morbidity/icd.html.
- 19. Levis, B. , Negeri Z., Sun Y., Benedetti A., Thombs B. D., and DEPRESSion Screening Data EPDS Group . 2020. “Accuracy of the Edinburgh Postnatal Depression Scale (EPDS) for Screening to Detect Major Depression Among Pregnant and Postpartum Women: Systematic Review and Meta‐Analysis of Individual Participant Data.” BMJ 371: m4022. 10.1136/bmj.m4022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Yee, L. M. , Martinez N. G., Nguyen A. T., Hajjar N., Chen M. J., and Simon M. A.. 2017. “Using a Patient Navigator to Improve Postpartum Care in an Urban Women's Health Clinic.” Obstetrics and Gynecology 129: 925–33. 10.1097/AOG.0000000000001977. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Rattan, J. , Leach J. M., Blanchard C., Tipre M., Bartlett T. R., Amiri A., Baskin M. L., Sinkey R., and Turan J. M.. 2025. “Health Insurance, Race, and Receipt of a Postpartum Visit Among Patients Giving Birth in a Referral Hospital in the US South.” Social Science & Medicine 372: 117922. 10.1016/j.socscimed.2025.117922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Steenland, M. W. , Oviedo D., Bates M. A., Zhou A., Zera C., Baicker K., and McConnell M. A.. 2025. “Effect of an Intensive Nurse Home Visiting Program on Postpartum Contraceptive Use and Birth Spacing: A Randomized Controlled Trial.” Obstetrics and Gynecology 145: 3–12. 10.1097/AOG.0000000000005786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Phillips, S. E. K. , Celi A. C., Margo J., Wehbe A., Karlage A., and Zera C. A.. 2024. “Improving Care Beyond Birth: A Qualitative Study of Postpartum Care After High‐Risk Pregnancy.” Journal of Women's Health 33: 1720–9. 10.1089/jwh.2024.0108. [DOI] [PubMed] [Google Scholar]
- 24. Kraus, A. C. , Quist‐Nelson J., Ryan S., Stuebe A., Young O. M., Volz E., Montiel C., Fiel L., Aktan I., and Tully K. P.. 2024. “Postpartum Care in a Cardio‐Obstetric Clinic After Preterm Preeclampsia: Patient and Healthcare Provider Perspectives.” American Journal of Obstetrics & Gynecology MFM 6: 101339. 10.1016/j.ajogmf.2024.101339. [DOI] [PubMed] [Google Scholar]
- 25. Stanhope, K. K. , Levinson A. N., Stallworth C. T., Leruth S., Clevenger E., Master M., Dunlop A. L., Boulet S. L., Jamieson D. J., and Blake S.. 2022. “A Qualitative Study of Perceptions, Strengths, and Opportunities in Cardiometabolic Risk Management During Pregnancy and Postpartum in a Georgia Safety‐Net Hospital, 2021.” Preventing Chronic Disease 19: E68. 10.5888/pcd19.220059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Vani, K. , Huang C., Hecht J., Lootens M. R., Karkowksy C. E., Fiori K., Allen E. G., and Lounsbury D. W.. 2024. “Implementing a Specialized Fourth Trimester Clinic Pilot for High‐Risk Individuals.” American Journal of Obstetrics & Gynecology MFM 6(12): 101523. 10.1016/j.ajogmf.2024.101523. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Brown, Z. , Messaoudi C., Silvia E., Bleau H., Meskill A., Flynn A., Abel‐Bey A. C., and Ball T. J.. 2023. “Postpartum Navigation Decreases Severe Maternal Morbidity Most Among Black Women.” American Journal of Obstetrics and Gynecology 229: 160.e1–e8. 10.1016/j.ajog.2023.01.002. [DOI] [PubMed] [Google Scholar]
- 28. Harris, S. A. , Eapen V., and Kohlhoff J.. 2024. “Implementing a National Navigation Service for Perinatal and Infant Mental Health: Early Learnings From the ForWhen Model.” Community Mental Health Journal 60: 581–8. 10.1007/s10597-023-01211-0. [DOI] [PubMed] [Google Scholar]
- 29. Nguyen, T. L. , Collins G. S., Lamy A., Devereaux P. J., Daurès J. P., Landais P., and Le Manach Y.. 2017. “Simple Randomization Did Not Protect Against Bias in Smaller Trials.” Journal of Clinical Epidemiology 84: 105–13. 10.1016/j.jclinepi.2017.02.010. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Supporting Information
Data Availability Statement
The data from this trial will be made available to researchers upon reasonable request to the corresponding author. Investigators will be required to submit a data analysis request form that requires elucidation of the specific data requested along with a statement that the investigator must provide: (1) a commitment to using the data only for research purposes and not to identify any individual participant, (2) a commitment to securing the data using appropriate computer technology, and (3) a commitment to destroying or returning the data after analyses are completed. Fully deidentified data will be provided after approval by the corresponding author and other members of the study team, as appropriate.