Abstract
Objectives: This study aimed to estimate the prevalence of diagnosed postpartum depression (PPD) and the likelihood of PPD among primiparous women. We also evaluated differences in the influence of various maternal factors associated with PPD in adolescent versus adult mothers. Methods: We conducted a retrospective cohort study using electronic health records linked to birth certificates to evaluate the associations between maternal factors and PPD diagnosis. The study population was stratified into adults and adolescents based on age at delivery. We evaluated socioeconomic, demographic, psychological, and clinical factors associated with PPD in each of the age-defined maternal cohorts using multivariable logistic regression analyses. Results: A total of 61,226 primiparous women, including 6435 (11%) mothers younger than 20 years old, were included in the study. The overall PPD rate was 4.0%, with the age-specific PPD rate measuring 1.6 times higher in adolescents than in adult women (6.1% vs. 3.8%). Compared with adults, adolescents were less likely to obtain firsttrimester prenatal care (33% vs. 16%), more likely to have recent tobacco use (11% vs. 6%), and more likely to have had an infection during pregnancy (5% vs. 1%). In adjusted models, significant factors for PPD in both groups included a history of depression or anxiety, tobacco use, and long-acting reversible contraception use. Conclusions: In this cohort of first-time mothers, adolescents had higher rates of PPD diagnosis as well as PPD-associated maternal factors than adults. Increased awareness of PPD risk in adolescents and early intervention, including integrating mental healthcare into prenatal care, may help benefit adolescents and reduce the risk and severity of PPD.
| Original language | English |
|---|---|
| Pages (from-to) | 218-228 |
| Number of pages | 11 |
| Journal | Birth |
| Volume | 51 |
| Issue number | 1 |
| DOIs | |
| State | Published - Mar 2024 |
| Externally published | Yes |
Funding
The data extraction for this work was funded in part by Bayer Healthcare (Grant Title: Long‐Acting Reversible Contraception and Preterm Birth in the Intermountain West: A Retrospective Review; Reference Number: RD‐OI‐0214). The authors received funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Developments, Office of Research on Women's Health (Grant number: K12HD085852). Jacqueline Kent‐Marvick received funding from the National Institute of Nursing Research of the National Institutes of Health (Award Number: T32NR013456). We thank the Pedigree and Population Resource of the Huntsman Cancer Institute, University of Utah (funded in part by the Huntsman Cancer Foundation) for its role in the ongoing collection, maintenance, and support of the Utah Population Database (UPDB). We also acknowledge partial support for the UPDB through grant P30 CA2014 from the National Cancer Institute, University of Utah, and from the University of Utah's Program in Personalized Health and Center for Clinical and Translational Science. We thank the University of Utah Center for Clinical and Translational Science (CCTS) (funded by NIH Clinical and Translational Science Awards), the Pedigree and Population Resource, the University of Utah Information Technology Services, and the Biomedical Informatics Core for establishing the Master Subject Index between the Utah Population Database, the University of Utah Health Sciences Center, and Intermountain Health Care. This research was also supported by the NCRR grant “Sharing Statewide Health Data for Genetic Research” (R01 RR021746, G. Mineau, PI), with additional support from the Utah State Department of Health and the University of Utah. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The data extraction for this work was funded in part by Bayer Healthcare (Grant Title: Long-Acting Reversible Contraception and Preterm Birth in the Intermountain West: A Retrospective Review; Reference Number: RD-OI-0214). The authors received funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Developments, Office of Research on Women's Health (Grant number: K12HD085852). Jacqueline Kent-Marvick received funding from the National Institute of Nursing Research of the National Institutes of Health (Award Number: T32NR013456). We thank the Pedigree and Population Resource of the Huntsman Cancer Institute, University of Utah (funded in part by the Huntsman Cancer Foundation) for its role in the ongoing collection, maintenance, and support of the Utah Population Database (UPDB). We also acknowledge partial support for the UPDB through grant P30 CA2014 from the National Cancer Institute, University of Utah, and from the University of Utah's Program in Personalized Health and Center for Clinical and Translational Science. We thank the University of Utah Center for Clinical and Translational Science (CCTS) (funded by NIH Clinical and Translational Science Awards), the Pedigree and Population Resource, the University of Utah Information Technology Services, and the Biomedical Informatics Core for establishing the Master Subject Index between the Utah Population Database, the University of Utah Health Sciences Center, and Intermountain Health Care. This research was also supported by the NCRR grant “Sharing Statewide Health Data for Genetic Research” (R01 RR021746, G. Mineau, PI), with additional support from the Utah State Department of Health and the University of Utah. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The data that support the findings of this study are available from the University of Utah. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of the University of Utah. Because this study aimed to explore differences in the association (direction and magnitude) of maternal factors for PPD between adolescent and adult mothers, the cohort was stratified and analyzed in two separate models depending on maternal age at first childbirth. Adolescents were defined as persons aged 12–19 years, and adults were defined as persons aged 20 years or older, according to the Centers for Disease Control and Prevention (CDC) definitions.31 Multivariable logistic regression was used to evaluate the associations between maternal factors and PPD, and estimates were reported as odds ratios (OR) with 95% confidence intervals (CIs). Analyses were run using SAS software 9.4 (Copyright, SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc.). The study was reviewed and approved by the University of Utah Institutional Review Board (IRB_00103461). This retrospective cohort study is part of a large project that linked three data sources at the patient level.21,22 The three data sources were (1) electronic health records (EHR) from Intermountain Healthcare, (2) EHR from the University of Utah Health, and (3) the Utah Population Database (UPDB). To illustrate, women who had their first live birth between 2008 and 2015 and at least one non-emergency visit within 10 years before their first live birth at either Intermountain Healthcare or the University of Utah Health were linked to Utah birth records available in the UPDB by probabilistic data linkage.21 This study was a secondary analysis of the established data and was designed to assess the relationship between LARC and pregnancy outcomes. The UPDB contains basic demographic information, birth records, death records, and US Census data on Utah residents.23 Using the linked data, we were able to use in- and outpatient EHR records as well as data available in UPDB for the same woman to measure study variables as described below. This retrospective cohort study is part of a large project that linked three data sources at the patient level.21,22 The three data sources were (1) electronic health records (EHR) from Intermountain Healthcare, (2) EHR from the University of Utah Health, and (3) the Utah Population Database (UPDB). To illustrate, women who had their first live birth between 2008 and 2015 and at least one non-emergency visit within 10 years before their first live birth at either Intermountain Healthcare or the University of Utah Health were linked to Utah birth records available in the UPDB by probabilistic data linkage.21 This study was a secondary analysis of the established data and was designed to assess the relationship between LARC and pregnancy outcomes. The UPDB contains basic demographic information, birth records, death records, and US Census data on Utah residents.23 Using the linked data, we were able to use in- and outpatient EHR records as well as data available in UPDB for the same woman to measure study variables as described below. Postpartum depression was defined as having a primary diagnosis code of major depression in at least one in- or outpatient encounter occurring within 1 year after the delivery date. The selection of depression-related diagnosis codes was based on Clinical Classification Software (CCS) provided by the Agency for Healthcare Research and Quality (AHRQ). Furthermore, we adapted the CCS groupings by adding additional diagnosis codes indicating depression from the “bipolar disorder” and “other miscellaneous mental conditions” categories to the “depressive disorder” category.24 The final list of diagnosis codes that we used to define postpartum depression is available in Table S1. We referred to a landmark systematic review to look for known PPD risk factors,8 and included additional factors that were determined a priori by the authors, who are clinical and/or content experts in maternal child health. Among the known and potential risk factors, we selected those that could be defined operationally in our data set using available data elements or relevant proxy measures. Demographic and socioeconomic factors included maternal age, race, ethnicity, educational attainment at delivery, residence at delivery (rural vs. urban), source of payment at delivery (measured by health-insurance type), and marital status at delivery.6–12 Psychological factors included pre-existing depression and anxiety.7,8,14 Finally, clinical factors included usage of prenatal care during the first trimester of pregnancy, tobacco use during pregnancy, long-acting reversible contraceptive (LARC) use 1 year before pregnancy, diabetes, hypertension, infections during pregnancy, cesarean delivery, excessive intrapartum bleeding, breastfeeding, preterm birth, and pre-pregnancy body mass index (BMI).6,12–18,25–27 The data sources and definitions used for the maternal factors are shown in Table S1. With respect to missing data, we replaced the missing value by an observed value using non-stochastic imputation techniques if missing data constituted less than 5% of the adolescent population.28,29 In brief, we used the value that accounted for the majority of the study population with the available data as a means of imputing the missing data value. If the missing data were 5% or more, we used the indicator method, which involved creating an additional “missing” category for the variable. Subsequently, the correlations between study variables were examined by correlation analysis.30 If any significant correlations were found, we considered combining all the correlated factors together to make a composite variable, or we retained only a variable with a large enough sample size and a better univariate association with PPD, excluding the others. Because this study aimed to explore differences in the association (direction and magnitude) of maternal factors for PPD between adolescent and adult mothers, the cohort was stratified and analyzed in two separate models depending on maternal age at first childbirth. Adolescents were defined as persons aged 12–19 years, and adults were defined as persons aged 20 years or older, according to the Centers for Disease Control and Prevention (CDC) definitions.31 Multivariable logistic regression was used to evaluate the associations between maternal factors and PPD, and estimates were reported as odds ratios (OR) with 95% confidence intervals (CIs). Analyses were run using SAS software 9.4 (Copyright, SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc.). The study was reviewed and approved by the University of Utah Institutional Review Board (IRB_00103461).
Keywords
- epidemiology
- postpartum depression
- risk factors
- teenage pregnancy