Abstract
BACKGROUND
The high maternal mortality and severe morbidity rates in the United States compared with other high-income countries have received national attention. Characterization of postpartum hospital readmissions within the first days after delivery hospitalization discharge could help to identify patients who need additional preparedness for discharge.
OBJECTIVE
This study aimed to investigate conditions at birth associated with postpartum readmissions occurring within 0 to 6 days and at 7 to 29 days after discharge from the delivery hospitalization.
STUDY DESIGN
We analyzed linked vital statistics and hospital discharge records of patients who gave birth in California during 2007 to 2018. We investigated hospital readmissions within 30 days after birth hospitalization discharge. We used multivariable logistic regression to investigate factors associated with early readmission (0–6 days) and later readmission (7–29 days) compared with no readmission within 30 days (reference). The risk factors assessed included maternal medical or obstetrical conditions before and at birth, birth hospitalization length of stay, and mode of delivery. Severe maternal morbidity was defined as the presence of any of the 21 indicators recommended by the Centers for Disease Control and Prevention.
RESULTS
Among 5,248,746 pregnant patients, 23,636 (0.45%) had an early postpartum readmission, whereas 24,712 (0.47%) had a later postpartum readmission. After adjustments, early readmission was most strongly associated with sepsis (adjusted odds ratio, 4.63; 95% confidence interval, 3.87–5.53), severe maternal morbidity (adjusted odds ratio, 3.46; 95% confidence interval, 3.28–3.65) at birth hospitalization, or preeclampsia before birth hospitalization (adjusted odds ratio, 3.67; 95% confidence interval, 3.54–3.81). The associations between later readmission and sepsis and severe maternal morbidity were similar, whereas the association between preeclampsia and later readmission was less strong (adjusted odds ratio, 1.65; 95% confidence interval, 1.57–1.73).
CONCLUSION
Pregnant patients with sepsis or severe maternal morbidity during delivery hospitalization or preeclampsia before birth hospitalization were at the highest risk for readmission within 6 days of discharge. These findings may be informative for efforts to improve postpartum care.
Key words: cesarean delivery, emergency department, hospitalization, major mental health condition, postpartum, preeclampsia, readmission, sepsis, severe maternal morbidity
AJOG Global Reports at a Glance.
Why was this study conducted?
The median time for postpartum readmission is 7 days after discharge, but risk factors for readmission within the immediate postpartum period (0–6 days) are not fully understood.
Key findings
Pregnant patients with preeclampsia, sepsis, or severe maternal morbidity during birth hospitalization had a 3- to four-fold increased odds of readmission within 6 days of discharge. Patients with preterm birth <34 weeks’ gestation and those with a major mental health condition had 2-fold increased odds of early readmission.
What does this add to what is known?
This study identified groups of patients who are at increased risk for readmission within 6 days of discharge following birth hospitalization. In particular, postpartum patients who have had sepsis, severe maternal morbidity at birth, or preeclampsia might benefit from proactive care planning.
The high maternal mortality and severe morbidity rates in the United States compared with other high-income countries have received national attention.1, 2, 3, 4 Maternal care quality during childbirth and within the immediate postpartum period are recognized as potential care improvement areas.5,6 To improve immediate postpartum care, the American College of Obstetricians and Gynecologists recommended in 2018 that all women should have contact with a maternal care provider within the first 3 weeks postpartum.6
More than half of maternal deaths in the United States occur during the postpartum period.4,5 In an effort to improve postpartum outcomes, information about the postpartum care utilization would be helpful. However, despite previous studies showing that the median time from discharge to postpartum readmission is 7 days (interquartile range, 2–19 days),7 the current literature on postpartum readmissions has focused mostly on factors associated with a readmission from 30 to 90 days postpartum.8, 9, 10, 11, 12 Few studies have investigated postpartum readmissions with ≤7 days of discharge.13 Identification of patients at risk of immediate return to the hospital after birth hospitalization could help to identify patients needing additional preparedness for discharge.
This study aimed to investigate conditions during birth hospitalization that were associated with postpartum readmissions occurring within the first week (0–6 days) after discharge.
Materials and Methods
Study population
We performed a cohort study using the California birth cohorts from 2007 to 2018. For these years, the California Department of Health Care Access and Information (formerly the Office of Statewide Health Planning and Development) linked the birth and fetal death certificate records with patient hospital discharge data during pregnancy and through the postpartum period.14 The birth and fetal death certificates include patient and pregnancy characteristics, and the discharge data include diagnosis and procedure codes from hospitalizations.
From 2007 to 2018, a total of 6,129,265 pregnant patients had a hospital delivery for a live birth or a fetal death after 20 weeks of gestation, including singleton and multifetal gestations. In the case of multifetal gestations, only 1 birth record was included in the analysis. After the exclusions, a total of 5,248,746 records were left for analysis (Figure 1). The Stanford University Institutional Review Board and the California State Committee for the Protection of Human Subjects reviewed and approved this study before its initiation.
Figure 1.
Flow diagram of the study cohort
BMI, body mass index.
Girsen. Early postpartum readmissions. Am J Obstet Gynecol Glob Rep 2022.
Postpartum readmissions
We identified postpartum readmissions based on patient discharge records from hospital inpatient readmissions that occurred after delivery hospitalization discharge. To ensure that we did not include hospital transfers as readmissions, we classified postpartum hospitalizations as transfers (not readmissions) for cases in which the (1) birth hospitalization discharge status was either within or to another hospital or the postpartum admission route was from within hospital or another hospital, and (2) time between postpartum admission and birth hospitalization discharge was ≤1 day. In the case of multiple postpartum hospitalizations for 1 patient, we selected the first occurrence for each.
We classified readmissions that occurred during 0 to 6 days after discharge from birth hospitalization as early and those that occurred during 7 to 29 days after discharge as later readmissions. These definitions were selected a priori by the investigators based on a multistate analysis showing that the median time between discharge and postpartum readmission is 7 days.7
Risk factors
We sought to evaluate conditions that were present before and during the birth hospitalization that could affect the probability of a readmission. Conditions were identified using the International Classification of Diseases, Ninth and Tenth Revision, Clinical Modification (ICD-9-CM and 1CD-10-CM) diagnosis codes.15 Conditions of interest occurring before birth hospitalizations included preeclampsia, preexisting diabetes mellitus, bleeding disorders, major mental health conditions, and thyroid disorders. Conditions or events of interest occurring during birth hospitalization included severe maternal morbidity, nontransfusion severe maternal morbidity, postpartum hemorrhage, sepsis, perineal trauma, cesarean delivery, stillbirth, preterm birth, and prolonged hospitalization. Details of the ICD-9-CM, ICD-10-CM diagnosis codes and others used to define the conditions analyzed are presented in Appendix 1. Severe maternal morbidity was defined as the presence of any of the 21 indicators recommended by the Centers for Disease Control and Prevention (CDC).16 Following the CDC recommendations, we also investigated nontransfusion severe maternal morbidity—cases in which blood products transfusion was not the sole indicator of a severe event. Preterm birth was defined as a birth <37 weeks of gestation as reported on the birth or fetal death certificate and was categorized as 34+0/7 to 36+6/7 or <34 weeks’ gestation. Prolonged length of delivery hospitalization was defined as ≥3 days for vaginal deliveries and ≥5 days for cesarean delivery.17
Confounders
We considered the following variables as potential confounders in multivariable models: maternal age at delivery, race/ethnicity, method of payment for delivery, education, parity, prepregnancy body mass index (BMI), multifetal gestation, preexisting diabetes mellitus and chronic hypertension. Prepregnancy BMI was calculated using the maternal weight and height (kg/m2). Preexisting diabetes mellitus and chronic hypertension were identified using the patient discharge data and the other variables were collected from the birth or fetal death certificate. These variables were selected a priori from available variables based on existing knowledge and causal diagrams.7,8,10,17
Causes of readmission
We also investigated the primary causes of early and later readmissions based on the primary ICD-9-CM and ICD-10-CM diagnosis code for the readmission (Appendix 2). We first categorized these into the 19 ICD-CM diagnosis code categories, which are largely organ specific.18 We then reclassified the diagnosis codes that were in the category of “Complications of Pregnancy, Childbirth, and the Puerperium” into the organ-specific categories as appropriate. Diagnoses in the perinatal category directly related to pregnancy and birth (eg, placental disorders) and nonspecific codes (eg, unspecified complication of pregnancy) were kept in the category.
Statistical analysis
We calculated the rate of early (0–6 days) and later (7–29 days) readmission per 1000 deliveries. We conducted a series of mixed effects logistic regression models with patients nested within delivery hospital to estimate odds ratios with 95% confidence intervals (CIs) for the association between each risk factor and (1) early readmission and (2) later readmission. Models were adjusted for variables as follows: (1) models for conditions occurring before birth hospitalization were adjusted for maternal age at delivery, race/ethnicity, method of payment for delivery, education, parity, prepregnancy BMI and multifetal gestation and (2) models for conditions occurring at birth hospitalization were additionally adjusted for diabetes mellitus and chronic hypertension or preeclampsia that existed before the birth hospitalization. Odds ratios approximated risk ratios because of the rarity of the outcomes. The level of significance was set at P<.05. Data management was done using SAS, version 9.4 (SAS Institute, Cary, NC) and data analysis using Stata/IP, version 15.1 (College Station, TX). The Strengthening the Reporting of Observational Studies in Epidemiology guidelines for cohort studies were followed.
Results
Patient characteristics
Included in the analysis were a total of 5,248,748 pregnant patients of whom 23,636 (0.45%) had a postpartum readmission at 0 to 6 days after birth hospitalization discharge (early readmission), and 24,712 (0.47%) were readmitted 7 to 29 days after discharge following birth hospitalization (later readmission) (Figure 1). The demographics and clinical characteristics of pregnant patients with early, later, and no readmission are presented in Table 1. Based on the frequencies observed, patients with early readmission were more likely to be Black, have college education, be nulliparous, have obesity, and have private insurance and a multifetal gestation than pregnant patients with no readmission within 30 days (Table 1). Patients with early readmission had higher frequencies of chronic hypertension and preexisting diabetes mellitus than patients without readmission within 30 days of birth hospitalization. Similarly, pregnant patients with a later readmission were more often Black or non-US born Hispanic, nulliparous, and more likely to have obesity, a multifetal gestation, preexisting diabetes mellitus, or chronic hypertension than patients without readmission. Patients with later readmission more often had public insurance and were less likely to have a college degree than those with no readmission (Table 1).
Table 1.
Demographics and clinical characteristics of pregnant patients with early (0–6 days after discharge) and later (7–29 days after discharge) postpartum readmission and pregnant patients with no postpartum readmission within 30 days in California from 2007 to 2018
| Characteristics | Early readmission 0–6 d after discharge from birth hospitalization n=23,636 | Later readmission 7–29 d after discharge from birth hospitalization n=24,712 | No readmission within 30 d of discharge from birth hospitalization n=5,200,398 |
|---|---|---|---|
| Column % | Column % | Column % | |
| Maternal age (y) | 29.9±6.6 | 28.3±6.6 | 28.9±6.2 |
| Race/ethnicity | |||
| US-born Hispanic | 6521 (27.6) | 8313 (33.6) | 1,373,902 (26.4) |
| Foreign-born Hispanic | 3654 (15.5) | 4662 (18.9) | 1,267,969 (24.4) |
| White, non-Hispanic | 7254 (30.7) | 6752 (27.3) | 1,480,949 (28.5) |
| Asian or Pacific Islander, non-Hispanic | 3271 (13.8) | 2584 (10.5) | 760,926 (14.6) |
| Black, non-Hispanic | 2783 (11.8) | 2181 (8.8) | 290,057 (5.6) |
| Other | 153 (0.7) | 220 (0.9) | 26,595 (0.5) |
| Insurance type for delivery | |||
| Public (Medi-Cal or other government) | 10,593 (44.8) | 13,466 (54.4) | 2,475,394 (47.6) |
| Private | 12,434 (52.6) | 10,625 (43.0) | 2,508,449 (48.2) |
| Self-pay or other | 609 (2.6) | 641 (2.6) | 216,555 (4.2) |
| Maternal education | |||
| No high school diploma | 3588 (15.2) | 4965 (20.1) | 1,047,059 (20.1) |
| High school graduate | 6050 (25.6) | 7239 (29.3) | 1,343,721 (25.8) |
| Associate degree or some college | 6979 (29.5) | 7149 (28.9) | 1,329,026 (25.6) |
| College graduate | 7019 (29.7) | 5359 (21.7) | 1,480,592 (28.5) |
| Nulliparous | 11,189 (47.3) | 11,025 (44.6) | 2,038,601 (39.2) |
| BMI category at prepregnancya | |||
| Underweight | 653 (2.8) | 740 (3.0) | 205,409 (4.0) |
| Normal weight | 9157 (38.7) | 9117 (36.9) | 2,488,954 (47.9) |
| Overweight | 6123 (25.9) | 6608 (26.7) | 1,357,961 (26.1) |
| Obese 1 | 3786 (16.0) | 4199 (17.0) | 684,930 (13.2) |
| Obese 2 | 2071 (8.8) | 2243 (9.1) | 290,262 (5.6) |
| Obese 3 | 1846 (7.8) | 1805 (7.3) | 172,882 (3.3) |
| Multifetal gestation | 921 (3.9) | 713 (2.9) | 84,263 (1.6) |
| Preexisting diabetes mellitus | 669 (2.8) | 646 (2.6) | 67,304 (1.3) |
| Chronic hypertension | 1315 (5.6) | 819 (3.3) | 73,265 (1.4) |
The data are presented for a total of 5,248,746 pregnancies.
BMI, body mass index.
Obese 1: BMI of 30 to <35; Obese 2: BMI of 35 to <40; Obese 3: BMI of ≥40.
Girsen. Early postpartum readmissions. Am J Obstet Gynecol Glob Rep 2022.
Conditions before birth hospitalization
After adjustments, preeclampsia had the strongest association with early readmission among the conditions present before birth hospitalization (adjusted odds ratio [aOR], 3.67; 95% CI, 3.54–3.81) (Table 2). The association between preeclampsia and later readmission was less strong (aOR, 1.65; 95% CI, 1.57–1.73). Other prebirth hospitalization factors associated with early readmission were a major mental health condition (aOR, 2.01; 95% CI, 1.91–2.12), bleeding disorder (aOR, 1.59; 95% CI, 1.46–1.74), preexisting diabetes mellitus (aOR, 1.73; 95% CI, 1.60–1.88), and thyroid disorder (aOR, 1.40; 95% CI, 1.32–1.48). These associations were similar for later readmission (Table 2).
Table 2.
Conditions before birth hospitalization associated with early readmission (0–6 days after discharge) and later readmission (7–29 days after discharge) compared with no readmission within 30 days of discharge from birth hospital in California from 2007 to 2018
| Conditions before birth hospitalizationa | Early readmission 0–6 d after discharge from birth hospitalization n=23,636 | Later Readmission 7–29 d after discharge from birth hospitalization n=24,712 | No readmission within 30 d of discharge from birth hospitalization n=5,200,398 | aOR (95% CI) for early readmission within 0–6 d | aOR (95% CI) for later readmission within 7–29 d |
|---|---|---|---|---|---|
| n (column %) | n (column %) | n (column %) | |||
| Preeclampsia | 3756 (15.9) | 1831 (7.4) | 188,702 (3.6) | 3.67 (3.54–3.81) | 1.65 (1.57–1.73) |
| Preexisting diabetes mellitus | 669 (2.8) | 646 (2.6) | 67,304 (1.3) | 1.73 (1.60–1.88) | 1.85 (1.71–2.01) |
| Bleeding disorder | 510 (2.2) | 420 (1.7) | 60,934 (1.2) | 1.59 (1.46–1.74) | 1.47 (1.33–1.62) |
| Major mental health condition | 1610 (6.8) | 1309 (5.3) | 124,385 (2.4) | 2.01 (1.91–2.12) | 1.95 (1.84–2.07) |
| Thyroid disorder | 1294 (5.5) | 993 (4.0) | 158,303 (3.0) | 1.40 (1.32–1.48) | 1.31 (1.23–1.40) |
The data are displayed for 5,248,746 pregnancies.
aOR, adjusted odds ratio; BMI, body mass index; CI, confidence interval.
Models were adjusted for maternal age, parity, prepregnancy BMI, education, race/ethnicity, and expected payment method for delivery.
Girsen. Early postpartum readmissions. Am J Obstet Gynecol Glob Rep 2022.
Conditions at birth hospitalization
Among conditions or events that occurred during the birth hospitalization, we found that sepsis had the strongest association with early readmission (aOR, 4.63; 95% CI, 3.87–5.53) and later readmission (aOR, 4.92; 95% CI, 4.08–5.93). Severe maternal morbidity at birth hospitalization significantly increased the odds of an early readmission (aOR, 3.46; 95% CI, 3.28–3.65), and the association was similar for nontransfusion only severe maternal morbidity (aOR, 3.71; 95% CI, 3.45–3.98). The odds of later readmission were also increased with severe maternal morbidity (aOR, 3.27; 95% CI, 3.09–3.46) and nontransfusion severe maternal morbidity at birth (aOR, 3.51; 95% CI, 3.24–3.81) (Table 3). Other notable birth hospitalization factors associated with early and later readmission were preterm birth <34 weeks’ gestation, cesarean delivery, postpartum hemorrhage, and prolonged birth hospitalization (Table 3). There were more stillbirths among patients with early (0.8%) and later (0.7%) readmission than among patients with no postpartum readmission (0.3%). However, the group sizes were too small to calculate point estimates for the associations. We found no significant association between perineal trauma and early or later readmission (Table 3).
Table 3.
Conditions at birth hospitalization associated with early readmission (0–6 days after discharge) and later readmission (7–29 days after discharge) compared with no readmission within 30 days of discharge from birth hospital in California from 2007 to 2018
| Conditions or events during birth hospitalizationa | Early readmission 0–6 d after discharge from birth hospitalization n=23,636 | Later Readmission 7–29 d after discharge from birth hospitalization n=24,712 | No readmission within 30 d of discharge from birth hospitalization n=5,200,398 | aOR (95% CI) for early readmission within 0–6 d | aOR (95% CI) for later readmission within 7–29 d |
|---|---|---|---|---|---|
| n (column %) | n (column %) | n (column %) | |||
| Severe maternal morbidity | 1614 (6.8) | 1296 (5.2) | 75,017 (1.4) | 3.46 (3.28–3.65) | 3.27 (3.09–3.46) |
| Severe maternal morbidity excluding transfusion-only cases | 816 (3.5) | 620 (2.5) | 31,508 (0.6) | 3.71 (3.45–3.98) | 3.51 (3.24–3.81) |
| Postpartum hemorrhage | 1593 (6.7) | 1456 (5.9) | 160,895 (3.1) | 1.69 (1.60–1.78) | 1.80 (1.71–1.90) |
| Sepsis | 131 (0.6) | 116 (0.5) | 4086 (0.1) | 4.63 (3.87–5.53) | 4.92 (4.08–5.93) |
| Perineal trauma | 974 (4.1) | 991 (4.0) | 212,459 (4.1) | 0.96 (0.90–1.02) | 0.98 (0.92–1.05) |
| Cesarean delivery | 12,416 (52.5) | 11,020 (44.6) | 1,677,636 (32.3) | 1.97 (1.92–2.02) | 1.55 (1.51–1.59) |
| Prolonged birth hospital length of stayb | 8508 (36.0) | 7240 (29.3) | 1,094,070 (21.0) | 1.56 (1.52–1.61) | 1.34 (1.30–1.38) |
| Stillbirth at ≥20 wk gestation | 184 (0.8) | 163 (0.7) | 17,815 (0.3) | —c | —c |
| Preterm birth: | |||||
| <34 wk | 1542 (6.5) | 1181 (4.8) | 109,132 (2.1) | 2.12 (2.00–2.23) | 1.94 (1.83–2.07) |
| 34–36 wk | 2531 (10.7) | 2177 (8.8) | 298,724 (5.7) | 1.52 (1.45–1.58) | 1.45 (1.38–1.51) |
| ≥37 wk | 19,563 (82.8) | 21,354 (86.4) | 4,792,542 (92.2) | Ref | Ref |
The data are displayed for 5,248,746 pregnancies.
aOR, adjusted odds ratio; BMI, body mass index; CI, confidence interval.
Models adjusted for maternal age, parity, prepregnancy BMI, education, race/ethnicity, and expected payment method for delivery, diabetes mellitus, chronic hypertension, and preeclampsia before birth hospitalization
Defined as ≥3 days for vaginal deliveries and ≥5 days for cesarean deliveries
Sample size too small for point estimate calculation.
Girsen. Early postpartum readmissions. Am J Obstet Gynecol Glob Rep 2022.
The 3 most common primary diagnoses for early readmission within 0 to 6 days from discharge were perinatal (Complications Of Pregnancy, Childbirth, And The Puerperium) (35% of early readmissions; n=7481), diagnoses related to the circulatory system (Diseases Of The Circulatory System) (29%; n=6278), and infectious and related complications (Infectious And Parasitic Diseases) (24%; n=5055), accounting for 88% of all readmissions in aggregate. The top 3 primary diagnoses for readmission during 7 to 29 days after discharge were slightly different with perinatal diagnoses (36% of later readmissions) being the top cause, followed by infectious (21%) and digestive (13%) diagnoses, accounting for 69% of readmissions in aggregate (Figure 2).
Figure 2.
Primary diagnosis for readmission
Primary diagnosis for readmission within 0–6 days (blue) and 7–29 days (red) after discharge from birth hospitalization.
Girsen. Early postpartum readmissions. Am J Obstet Gynecol Glob Rep 2022.
Comment
Principal findings
We found a more than 4-fold higher risk for early postpartum readmission (within 6 days of discharge) among patients who had sepsis during birth hospitalization and a >3-fold higher risk for early postpartum readmission among patients who had severe maternal morbidity or preeclampsia. In addition, there was a 2-fold higher risk for early readmission among patients with preterm birth <34 weeks’ gestation or a major mental health condition. Associations between conditions and later readmissions (7–29 days after discharge) were overall less pronounced with the main difference being for preeclampsia, which had a much stronger association with early readmissions than with later readmission. The large majority (88%) of readmissions during the first week were because of perinatal, circulatory, and infectious conditions. The causes for later readmission were more dispersed overall with perinatal, infectious, and digestive causes covering more than half (69%) of the later readmissions.
Results in the context of what is known
Maternal care during the immediate postpartum period has received significant attention.6,19 Based on the United States vital statistics records from 2011 to 2015, 18.6% of pregnancy-related mortalities occurred within 1 to 6 days postpartum and 21.4% occurred within 7 to 42 days postpartum.5 Furthermore, we have shown previously that approximately 12% of severe maternal morbidity cases occurred after delivery hospitalization discharge.20 Despite a median time from discharge to readmission of 7 days,7 previous studies have mainly investigated readmissions within 30 or 60 days of discharge after delivery hospitalization8, 9, 10,21 with only a few examining readmissions within the first 7 days.13 We selected postpartum readmission within the first 6 days as our primary outcome as a vehicle to examine the care needed during the immediate postpartum period. Although postpartum readmission as a quality metric of obstetrical care is a controversial topic (Combs AC 2022), unarguably, return to the hospital within days after discharge is an untoward outcome with potential negative impact on the wellbeing of the mother and the infant. We found that the strongest predictors of early readmission were sepsis, severe maternal morbidity, and preeclampsia. Our findings are in line with a recent study by DeClercq et al22 who showed that sepsis cases contribute most to the severe maternal morbidity events beyond delivery hospitalization. In addition, previous studies have shown that severe maternal morbidity at delivery is associated with an increased risk for readmission, although these studies did not focus on early readmissions.10,23 Our ability to differentiate the risk factors for early and later readmission revealed how the effect of sepsis and severe maternal morbidity on readmission may persist past the early postpartum period, whereas preeclampsia seems to be much more of a risk factor in the early postpartum period; preeclampsia increased the odds of early readmission by almost 4-fold, whereas the odds of later readmission were increased only about 1.5-fold.
In this study, the most common diagnosis for readmission within 0 to 6 days was perinatal conditions (complications of pregnancy, childbirth), followed by circulatory systems conditions (including hypertensive disorder) and infectious conditions. Our findings are in line with previous literature demonstrating that infection,7,24,25 hypertensive disorders,7,26,27 and psychiatric illness7 are most common indicators for postpartum readmission. Interestingly, the primary diagnoses for readmission between 7 to 29 days after discharge were slightly different from those in the first week with perinatal, infectious, and digestive conditions accounting for 69% of the diagnoses and circulatory causes being less frequently diagnosed than during the early readmission. This is aligned with previous literature showing that a high percentage of postpartum readmissions within the first 6 weeks of delivery are for pneumonia (infectious), appendicitis, and cholecystitis (digestive causes).25 Furthermore, our results support the previous findings that hypertensive readmissions are more likely to occur within the first few days28 and that proactive monitoring and treatment of these patients before discharge are warranted.
Clinical implications
Because so many of the maternal morbidity and mortality cases are tied to the postpartum period, understanding postpartum inpatient and outpatient care utilization is important in improving maternal outcomes. Our results suggest that pregnant patients requiring readmission in the immediate postpartum period have similar risk factors. The strongest risk factors for readmission within 6 days of discharge in our study were sepsis, severe maternal morbidity, and preeclampsia. Not surprisingly, these are all aligned with the leading causes of pregnancy-related mortality in California, which are cardiovascular disease, sepsis or infection, hemorrhage, and hypertensive disorders.29 Patients with early readmission had the highest frequency of prolonged birth hospital length of stay, thus suggesting that their conditions were also the most severe. It is unclear whether incorporation of immediate postpartum check-ins or treat-and-discharge visits in an outpatient setting would be beneficial or more appropriate for some of the postpartum patients. Preeclampsia was specifically associated with early readmission as opposed to later readmission, which may be improved by better planning in terms of antihypertensive medications or care team follow-up right after the hospital discharge as suggested by others.28 Similar to hypertensive disorders, there might be opportunities to improve discharge readiness of patients with sepsis during birth hospitalization, for example, by proactive treatment with antibiotics and close postpartum follow-up. Furthermore, major mental health conditions before birth hospitalization contributed about a 2-fold increase in the odds of both early and later postpartum readmission, thus emphasizing the role of mental health conditions on the overall maternal health during the immediate postpartum period.
Research implications
The results of this study highlight that postpartum patients with sepsis, severe maternal morbidity, or preeclampsia are at highest risk for immediate return to the hospital. It is unclear how much the recent national guidelines on postpartum care6 have improved readmission outcomes and if postpartum readmission is surrogate for severe maternal morbidity or if it actually improves outcomes. Furthermore, more research on the effects of birth hospital discharge planning would be needed. The nonobstetrical literature has shown that discharge education (giving instructions, assessing patient understanding, reinstruction, and reassessing patient understanding) has a potential to decrease readmission rates after heart failure.30 In addition, hospital discharge quality (defined as patient experience of having adequate information and resources at the time of discharge) has been associated with lower readmission rates among a nonobstetrical population.31 A recent study on hypertensive postpartum readmissions suggests that proactive treatment of postpartum hypertension before discharge could significantly reduce readmissions and improve outcomes.28 Finally, an evaluation of the systems of care associated with early readmissions would be a key addition to this study.
Strengths and limitations
Among our study's limitations, we acknowledge that the data were derived from linked patient discharge records and birth certificate data and are hence subject to miscoding and underreported information.32 In addition, some patient comorbidities may not have been accounted for in our data on delivery hospitalizations.33 Our data set did not include information on the parameters of clinical management including medications, vital signs, laboratory values, or the severity of conditions, all of which could have impacted the risk for postpartum readmission. Although the California population is diverse and large (accounting for 1 in 8 births in the United States), we acknowledge that examining data from just 1 state might affect the generalizability of our results.
The strengths of our study include a large and diverse population-based cohort that was used for the analysis. The hospitalizations in our data set were linked across pregnancy and the postpartum period and linked to both live birth and fetal death certificates. As such, our data set is a highly unusual source of data to investigate these important but rare events. The size of our dat aset enabled us to examine rare risk factors like sepsis or stillbirth. Furthermore, we were able to examine primary diagnoses and account for patient characteristics in the analysis.
Conclusion
Our study found that pregnant patients with sepsis and severe maternal morbidity had the highest risk for readmission either within 0 to 6 days (early) or 7 to 29 days (later) after discharge from birth hospitalization. Preeclampsia was a major risk factor for early but not for later readmission. The vast majority of early readmissions were owing to perinatal, circulatory, and infectious causes, whereas causes for later readmissions were more dispersed and associated with infectious and digestive conditions to a greater extent. Future studies should investigate the impact of proactive management and short-term postpartum follow-up of pregnant patients with these conditions.
Acknowledgments
The authors would like to acknowledge Suzan Carmichael, PhD and Henry Lee, MD and the California Perinatal Quality Care Collaborative for providing the study data and the Dunlevie Maternal-Fetal Medicine Center for Discovery, Innovation and Clinical Impact at Stanford University for supporting this work.
Footnotes
The authors report no conflict of interest.
This study was supported by the Dunlevie Maternal-Fetal Medicine Center for Discovery, Innovation and Clinical Impact at Stanford University.
This study was presented as a poster at the 40th annual meeting of the Society for Maternal-Fetal Medicine, Grapevine, TX, February 3–8, 2020.
Patient consent was not required because no personal health information or identifiers were included in the data set.
Cite this article as: Girsen AI, Leonard SA, Butwick AJ, et al. Early postpartum readmissions: identifying risk factors at birth hospitalization. Am J Obstet Gynecol Glob Rep 2022;2:100094.
References
- 1.Callaghan WM, Creanga AA, Kuklina EV. Severe maternal morbidity among delivery and postpartum hospitalizations in the United States. Obstet Gynecol. 2012;120:1029–1036. doi: 10.1097/aog.0b013e31826d60c5. [DOI] [PubMed] [Google Scholar]
- 2.Creanga AA, Berg CJ, Syverson C, Seed K, Bruce FC, Callaghan WM. Pregnancy-related mortality in the United States, 2006-2010. Obstet Gynecol. 2015;125:5–12. doi: 10.1097/AOG.0000000000000564. [DOI] [PubMed] [Google Scholar]
- 3.MacDorman MF, Declercq E, Cabral H, Morton C. Recent increases in the U.S. Maternal mortality rate: disentangling trends from measurement issues. Obstet Gynecol. 2016;128:447–455. doi: 10.1097/AOG.0000000000001556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tikkanen R, Gunja MZ, FitzGerald M, Zephyrin L. Maternal mortality and maternity care in the United States compared to 10 other developed countries. 2020. Available at:https://www.commonwealthfund.org/publications/issue-briefs/2020/nov/maternal-mortality-maternity-care-us-compared-10-countries. Accessed August 10, 2022.
- 5.Petersen EE, Davis NL, Goodman D, et al. Vital Signs: pregnancy-related deaths, United States, 2011-2015, and strategies for prevention, 13 states, 2013-2017. MMWR Morb Mortal Wkly Rep. 2019;68:423–429. doi: 10.15585/mmwr.mm6818e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.ACOG Committee Opinion No. 736: optimizing postpartum care. Obstet Gynecol. 2018;131:e140–e150. doi: 10.1097/AOG.0000000000002633. [DOI] [PubMed] [Google Scholar]
- 7.Clapp MA, Little SE, Zheng J, Robinson JN. A multi-state analysis of postpartum readmissions in the United States. Am J Obstet Gynecol. 2016;215:113. doi: 10.1016/j.ajog.2016.01.174. e1–10. [DOI] [PubMed] [Google Scholar]
- 8.Zuckerman RB, Sheingold SH, Orav EJ, Ruhter J, Epstein AM. Readmissions, observation, and the hospital readmissions reduction program. N Engl J Med. 2016;374:1543–1551. doi: 10.1056/NEJMsa1513024. [DOI] [PubMed] [Google Scholar]
- 9.Fein A, Wen T, Wright JD, et al. Postpartum hemorrhage and risk for postpartum readmission. J Matern Fetal Neonatal Med. 2021;34:187–194. doi: 10.1080/14767058.2019.1601697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Nam JY, Park EC. The relationship between severe maternal morbidity and a risk of postpartum readmission among Korean women: a nationwide population-based cohort study. BMC Pregnancy Childbirth. 2020;20:148. doi: 10.1186/s12884-020-2820-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Batra P, Fridman M, Leng M, Gregory KD. Emergency department care in the postpartum period: California births, 2009-2011. Obstet Gynecol. 2017;130:1073–1081. doi: 10.1097/AOG.0000000000002269. [DOI] [PubMed] [Google Scholar]
- 12.Miller EC, Wen T, Elkind MSV, Friedman AM, Boehme AK. Infection During delivery hospitalization and risk of readmission for postpartum stroke. Stroke. 2019;50:2685–2691. doi: 10.1161/STROKEAHA.119.025970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Matas JL, Mitchell LE, Sharma SV, Louis JM, Salemi JL. Severe maternal morbidity at delivery and postpartum readmission in the United States. Paediatr Perinat Epidemiol. 2021;35:627–634. doi: 10.1111/ppe.12762. [DOI] [PubMed] [Google Scholar]
- 14.The California Department of Health care access and information (HCAI). Research data request information. 2021. Available at:https://hcai.ca.gov/data-and-reports/research-data-request-information/. Accessed February 27, 2022.
- 15.Leonard SA, Kennedy CJ, Carmichael SL, Lyell DJ, Main EK. An expanded obstetric comorbidity scoring system for predicting severe maternal morbidity. Obstet Gynecol. 2020;136:440–449. doi: 10.1097/AOG.0000000000004022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Centers for Disease Control and Prevention. Severe maternal morbidity in the United States. 2021. Available at:https://www.cdc.gov/reproductivehealth/maternalinfanthealth/severematernalmorbidity.html#anchor_References. Accessed February 27, 2022.
- 17.Hebert PR, Reed G, Entman SS, Mitchel EF, Berg C, Griffin MR. Serious maternal morbidity after childbirth: prolonged hospital stays and readmissions. Obstet Gynecol. 1999;94:942–947. doi: 10.1016/s0029-7844(99)00419-6. [DOI] [PubMed] [Google Scholar]
- 18.Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project (HCUP). 2020. Available at:www.hcup-us.ahrq.gov/faststats/national/inpatientcommondiagnoses.jsp. Accessed February 27, 2022. [PubMed]
- 19.Society for Maternal-Fetal Medicine (SMFM) Quality and Safety and Health Policy Committees. Bailit JL, Gregory KD. Society for Maternal-Fetal Medicine (SMFM) Special Report: current approaches to measuring quality of care in obstetrics. Am J Obstet Gynecol. 2016;215:B8–16. doi: 10.1016/j.ajog.2016.06.048. [DOI] [PubMed] [Google Scholar]
- 20.Girsen AI, Sie L, Carmichael SL, et al. Rate and causes of severe maternal morbidity at readmission: California births in 2008-2012. J Perinatol. 2020;40:25–29. doi: 10.1038/s41372-019-0481-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wen T, Yu VX, Wright JD, et al. Postpartum length of stay and risk for readmission among women with preeclampsia. J Matern Fetal Neonatal Med. 2020;33:1086–1094. doi: 10.1080/14767058.2018.1514382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Declercq ER, Cabral HJ, Cui X, et al. Using longitudinally linked data to measure severe maternal morbidity. Obstet Gynecol. 2022;139:165–171. doi: 10.1097/AOG.0000000000004641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Black CM, Vesco KK, Mehta V, Ohman-Strickland P, Demissie K, Schneider D. Hospital readmission following delivery with and without severe maternal morbidity. J Womens Health (Larchmt) 2021;30:1736–1743. doi: 10.1089/jwh.2020.8815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Clapp MA, Robinson JN, Little SE. The relationship between the rising cesarean delivery and postpartum readmission rates. J Perinatol. 2017;37:355–359. doi: 10.1038/jp.2016.252. [DOI] [PubMed] [Google Scholar]
- 25.Belfort MA, Clark SL, Saade GR, et al. Hospital readmission after delivery: evidence for an increased incidence of nonurogenital infection in the immediate postpartum period. Am J Obstet Gynecol. 2010;202:35. doi: 10.1016/j.ajog.2009.08.029. e1–7. [DOI] [PubMed] [Google Scholar]
- 26.Mogos MF, Salemi JL, Spooner KK, McFarlin BL, Salihu HH. Hypertensive disorders of pregnancy and postpartum readmission in the United States: national surveillance of the revolving door. J Hypertens. 2018;36:608–618. doi: 10.1097/HJH.0000000000001594. [DOI] [PubMed] [Google Scholar]
- 27.Chornock R, Iqbal SN, Kawakita T. Racial disparity in postpartum readmission due to hypertension among women with pregnancy-associated hypertension. Am J Perinatol. 2021;38:1297–1302. doi: 10.1055/s-0040-1712530. [DOI] [PubMed] [Google Scholar]
- 28.Lovgren T, Connealy B, Yao R, Dahlke JD. Postpartum management of hypertension and effect on readmission rates. Am J Obstet Gynecol MFM. 2022;4 doi: 10.1016/j.ajogmf.2021.100517. [DOI] [PubMed] [Google Scholar]
- 29.California Pregnancy Mortality Surveillance System. California pregnancy-related deaths, 2008–2016. 2021. Available at:https://www.cdph.ca.gov/Programs/CFH/DMCAH/surveillance/CDPH%20Document%20Library/CA-PMSS/CA-PMSS-Surveillance-Report-2008-2016.pdf. Accessed June 6, 2022.
- 30.Oh EG, Lee HJ, Yang YL, Kim YM. Effectiveness of discharge education With the teach-back method on 30-day readmission: a systematic review. J Patient Saf. 2021;17:305–310. doi: 10.1097/PTS.0000000000000596. [DOI] [PubMed] [Google Scholar]
- 31.Henke RM, Karaca Z, Jackson P, Marder WD, Wong HS. Discharge planning and hospital readmissions. Med Care Res Rev. 2017;74:345–368. doi: 10.1177/1077558716647652. [DOI] [PubMed] [Google Scholar]
- 32.Lydon-Rochelle MT, Holt VL, Nelson JC, et al. Accuracy of reporting maternal in-hospital diagnoses and intrapartum procedures in Washington State linked birth records. Paediatr Perinat Epidemiol. 2005;19:460–471. doi: 10.1111/j.1365-3016.2005.00682.x. [DOI] [PubMed] [Google Scholar]
- 33.Salemi JL, Hansen MA, Modak S, et al. Estimating the obstetric co-morbidity burden using administrative data: the impact of the pregnancy-related assessment window. Paediatr Perinat Epidemiol. 2020;34:440–451. doi: 10.1111/ppe.12593. [DOI] [PubMed] [Google Scholar]