Analysis of Mood and Anxiety Disorder Content in Common U.S. Obstetrics and Gynecology Residency Educational Resources.

The primary outcome of this study was the percentage of total content present in OBGYN residency third-party educational resource materials related to depression and anxiety. Secondary outcomes included estimates of the percentage of content of depression and anxiety by reference source, percentage of content pertaining to perinatal vs nonperinatal (gynecologic) context, and antenatal vs postpartum contexts. We hypothesized that there would be a low percentage of depression and anxiety disorder content among OBGYN resident educational resources. We also hypothesized that content would most commonly contain perinatal topic areas compared with nonperinatal (gynecologic) areas and postpartum topics compared with antepartum topics. To explore these hypotheses, this study used an inductive approach to qualitative content analysis. 21 Because this study did not involve human subjects research, institutional review board approval was not necessary. The most recent editions of commonly used educational resources were included. These resources were identified based on the published results of a multi-institutional survey of OBGYN residents published in 2020 identifying the most commonly used educational resources for independent learning during OBGYN residency. 18 This initial inclusion list was composed of the American College of Obstetricians and Gynecologists (ACOG) Practice Bulletins (used by 48% of OBGYN residents for independent learning), ACOG Committee Opinions (used by 47%), Personal Review of Learning in Obstetrics and Gynecology (PROLOG) (used by 35%), OBGYN textbooks (used by 45%; Williams Obstetrics 22 and Gabbe’s Obstetrics: Normal & Problem Pregnancies 23 selected for inclusion in this study), Online Question Bank (used by 20%; TrueLearn OBGYN Practice Questions for Residents: CREOG In-Training Exam Question Bank selected for inclusion this study), and Society for Academic Specialists in General Obstetrics and Gynecology (SASGOG) Pearls of Exxcellence learning modules (used by 8%). Two references cited by the study were not included in the current study (faculty materials and the Core Clinical Cases in Obstetrics and Gynecology: A Problem-Solving Approach book 24 ). Faculty materials vary by institution and could not be systematically reviewed. Core Clinical Cases in Obstetrics and Gynecology: A Problem-Solving Approach 24 was last published in 2011 (9 years out of date at the time of our review) and used by only 2% of OBGYN residents for independent learning; therefore, this resource was omitted due to question of current relevance among residents. Subsequently, a cohort of OBGYN residents (n = 32) from a single academic institution in New York in April 2020 contributed to the authors’ decision to add additional references for inclusion (Dr. Wall’s OBGYN board prep course materials and CREOGs Over Coffee podcast series). This information was obtained informally in response to the open-ended question, “What study materials do you use to prepare for standardized OBGYN exams?” This question was posed to this group of individuals during an OBGYN resident meeting by the residency program director. Volunteered answers by the group were noted by one author, and nonoverlapping sources were considered by both authors before inclusion. Once these resources were selected, copies of relevant materials were accessed through open access Internet webpages or university library access. Online question bank subscription was provided previously by local graduate medical education funding for one of the authors; PROLOG and Dr. Wall’s materials had been previously purchased by one of the authors before this study. All data were accessed and archived on July 30, 2020, for future analyses. We independently hand-searched all questions contained in the online question bank, PROLOG books and Dr. Walls tests, chapter titles in textbooks, table of contents for practice bulletins and committee opinions, episode titles for podcast series audio files, and table of contents for titles of brief focused reviews. Seventeen keywords were initially prespecified based on the traditional classification systems for mental illness, 25 with additional keywords, synonyms, and related concepts added iteratively from scanning the references. Those with potential relevance were then reviewed in full by the 2 independent reviewers, and if content contained prespecified mental health topics, they were included. All references were reviewed in text form, except for podcasts, which were reviewed in audio file form. Any discrepancies in decisions to include a reference were discussed and adjudicated between reviewers. This adjudication process involved (1) both reviewers re-reading the reference in question in full, (2) discussion of reference topic in relation to prespecified coding hierarchy, and (3) consensus decision on whether the reference met inclusion criteria. Codes were developed by process of conventional content analysis. 21 Inductive content analysis was used in this case because, to our knowledge, no previous studies have dealt with the research topic and reference materials. 21 , 26 We used 2 sequential rounds of preliminary coding of a limited dataset to iteratively develop the coding dictionary, which contained coded categories derived directly from the text data. 26 Dedoose, version 9.0.17 (SocioCultural Research Consultants LLC, Los Angeles, California), a cloud application for managing, analyzing, and presenting qualitative and mixed-methods research data, was used to develop the coding dictionary. First, open coding was used by both reviewers independently, with categories freely generated by both reviewers independently working in Dedoose on the same limited dataset. The results were then discussed, allowing for identification of data requiring addition of new codes as well as identifying coding discrepancies and rectifying them by refining code definitions. Codes were then categorized and grouped under hierarchical headings, and definitions of each category, subcategory, and code were finalized. 26 Second, the revised coding scheme was again piloted with a limited dataset by both reviewers, with further revision and refinement of the coding scheme and code definitions before implementing the coding process to the whole dataset. Interrater reliability was not calculated before implementing the final coding scheme to the entire dataset. The coding for this study included 10 general categories and 35 subcategories of mental health–related topics, including psychological disorder type, epidemiology, risk factors, screening, diagnostic criteria, symptom recognition, pathophysiology, principles of treatment, psychopharmacology, perinatal safety, and related conditions. We also assigned 5 perinatal classifiers to each reference. The classifiers were obstetric (pregnant, postpartum, or lactating), nonobstetric (not pregnant, postpartum, or lactating), both obstetric and nonobstetric topic areas, antenatal obstetrics, or postnatal obstetrics. Content was categorized under heading levels (categories) and subheadings (subcategories) using an inductive approach that followed a typical textbook chapter outline format commonly used to teach clinical concepts to learners. 27 The classifiers were derived from the research questions we sought to answer related to mental health topic distribution in the specialty of OBGYN. Coding of each reference was then independently performed by 2 reviewers. The review process entailed reading the entire reference and applying the prespecified codes for the categories, subcategories, classifiers, and codes using Dedoose. We coded the binary outcome of the presence or absence of each category, subcategory, classifier, and code for each included resource. Although included in coding, for analysis, we specifically focused on mood and anxiety disorders and do not report here on content solely pertaining to substance-related disorders, feeding and eating disorders, psychological sexual dysfunction, intimate partner violence, sexual assault, or other disorders. Recognizing that these are important concepts in medical education and clinical practice for OBGYN specialists, these disorders were considered beyond the scope of the research question asked here. We rated the content in each category, subcategory, and code using the following ratings: 0 for absent content, 1 for minimal content, and 2 for helpful content and topic well discussed. A rating of 1 was received for any mention of the topic, whereas a rating of 2 was assigned if the content provided helpful information such as could be applied by the reader in a clinical context. This rating was performed in a similar manner to that described for the coding procedures above. We used 2 sequential rounds of preliminary ratings of a limited dataset to develop a unified rating approach. First, open rating was used by both reviewers independently working in Dedoose on the same limited dataset. The results were then discussed, identifying rating discrepancies between reviewers and rectifying them by refining rating definitions. Second, the rating scheme was again piloted with a limited dataset by both reviewers, with further revision and refinement of the rating scheme and definitions before implementing the coding process to the whole dataset. Interrater reliability was not calculated before implementing the final ratings to the entire dataset. Descriptive statistics were used to determine percentages of content categories, subcategories, and classifiers. We calculated a denominator for each educational resource as outlined in Table 1 . The percentage of content containing depression and anxiety content was calculated by (1) dividing the number of question items discussing these topics by the total number of questions contained in the question bank (PROLOG, Dr. Wall’s OBGYN board prep course materials, and TrueLearn OBGYN Practice Questions for Residents), (2) dividing the total number of chapters discussing these topics by the total number of chapters contained in the textbook ( Williams Obstetrics 22 and Gabbe’s Obstetrics: Normal & Problem Pregnancies 23 ), (3) dividing the number of brief focused reviews discussing these topics by the total number of brief focused reviews contained in the collection (SASGOG Pearls of Exxcellence learning modules and CREOGs Over Coffee podcast series), and (4) dividing the number of practice bulletins (or committee opinions) discussing these topics by the total number of published practice bulletins (or committee opinions). 28 – 39 The κ statistical test was used to determine interrater reliability for the 10 categories, 35 subcategories, and 5 perinatal classifiers and for reviewer ratings of content in each domain from 0 to 2. An a priori interrater reliability of 90% was set, and any disagreement of rater coding and content rating was adjudicated before final analysis. All statistical tests were performed using SPSS Statistics, version 29.0.0.0 (IBM Inc., Armonk, New York).

From the 7 total reference materials chosen for this review, the total number of items coded for relevant mental health keywords and included in the content analysis was 36 (n = 8 of 1,008 TrueLearn OBGYN Practice Questions for Residents, n = 5 of 600 Dr. Wall’s OBGYN board prep course materials, n = 7 of 871 PROLOG questions, n = 2 of 124 textbook cha pters, n = 12 of 326 ACOG clinical guidance documents, n = 1 of 127 SASGOG Pearls of Exxcellence learning modules, and n = 1 of 166 CREOGs Over Coffee podcasts) 28 – 47 ( Table 1 ). The total number of coded mental health categories, subcategories, and classifiers across these 36 sources was 91. Interrater reliability for all 91 assigned categories, subcategories, and classifiers was κ = 0.835 (95% CI, 0.804–0.866) and increased to κ = 1.00 after adjudication between reviewers. The percentage of content containing learning domains of mood and anxiety disorders across all independent learning educational resources was 1.1% ( Table 1 and Figure 1 ). The ACOG Practice Bulletins contained the highest percentage of this content at 5.0% (95% CI, 1.4%–12.3%), and the CREOGs Over Coffee podcast contained the lowest at 0.6% (95% CI, 0.02%–3.3%) ( Table 1 and Figure 1 ). Resources containing the highest percentage of total content in mental health topics also corresponded to resources reported to be most frequently used by OBGYN residents ( Table 1 ). The mood and anxiety disorder content contained in the included reference materials varied considerably. We found that among all references, a depression code was applied most frequently (30 [83.3%]), followed by an anxiety code (12 [33.3%]) and bipolar disorder code (9 [25.0%]) ( Table 2 ). All 7 reference materials included had a depression code applied, whereas only 3 had any anxiety code and 4 had any bipolar code applied. Few codes were applied for content pertaining to obsessive-compulsive disorder (3 [8.3%]), posttraumatic stress disorder (5 [13.9%]), schizophrenia (2 [5.6%]), or postpartum psychosis (4 [11.1%]). The TrueLearn OBGYN Practice Questions for Residents 45 and the ACOG Practice Bulletins 36 – 39 contained the greatest breadth of content covering different mental health diagnoses. Some references had limited breadth of content; for example, the SASGOG Pearls of Exxcellence learning modules 47 had 1 topic review on premenstrual dysphoric disorder, and the CREOGs Over Coffee podcast 46 had 1 episode on perinatal depression. Principles of treating mental health conditions, in particular, pharmacotherapy topics, were well represented among the reviewed reference materials (see Supplemental Digital Appendix 3 at http://links.lww.com/ACADMED/B663 ). Selective serotonin reuptake inhibitors were the most common included pharmacotherapy content overall, particularly among practice test questions, with 6 of 7 PROLOG questions and 4 of 8 TrueLearn questions testing on them, demonstrating the topic’s significant overrepresentation. Content on psychiatric medications with known teratogenicity, such as lithium and valproic acid, was present in 3 reference types reviewed, inclusive of a total of 2 test questions, both textbook chapters, and 3 ACOG publications. Lactation safety of psychiatric medications had the lowest percentage of content (8.3%). The reference materials primarily contained content coded in a gynecologic context followed by a postpartum context and more rarely an antepartum context ( Table 2 ). Some reference materials endeavored to balance discussion of mental health topics in both gynecologic and perinatal contexts, for example, textbook chapters and ACOG Committee Opinions. There was, however, an overall overrepresentation of gynecologic content, particularly in questions banks, Pearls of Exxcellence learning modules, and ACOG Practice Bulletins (see Supplemental Digital Appendix 4 at http://links.lww.com/ACADMED/B663 ). Among all resources, content areas demonstrating consistently high assigned rating (i.e., rating of 2, reflecting helpful information such as could be applied by the reader in a clinical context) included depressive disorder, principles of treatment, and pharmacotherapy ( Table 3 ). The content ratings for schizophrenia, postpartum psychosis, posttraumatic stress disorder, and obsessive-compulsive disorder were overall low ( Table 3 ). The pathophysiology of mental health conditions, lactation safety, and fetal and maternal adverse effects of psychiatric medications and of psychiatric disease also had very low frequencies of a content rating of 2 ( Table 3 ). Interrater reliability for assigned content ratings (range, 0–2) was κ = 0.970 (95% CI, 0.956–0.984). No individual educational resource reviewed in this study contained all recommended CREOG OBGYN residency curriculum development topics related to mental health conditions across the reproductive spectrum (see Supplemental Digital Appendix 1 at http://links.lww.com/ACADMED/B663 ).

This content analysis of commonly used educational resources for independent learning and annual standardized exam preparation for U.S. OBGYN residents demonstrates extremely low composition of mood and anxiety disorder topics with concurrent low content quality of included information. Of the reviewed resources, no one individual resource contains information in all the mental health–related content areas recommended by CREOG, which underscores the need for a comprehensive and evidence-based curriculum to prepare OBGYN residents for clinical practice after residency. Of all reviewed sources, this study shows that Committee Opinions and Practice Bulletins published by ACOG contain the highest total percentage of content related to mental health conditions. Fortunately, these resources are also reported to be most commonly used by OBGYN residents for independent learning. 18 Since this content analysis was performed, 2 additional comprehensive ACOG clinical guidance documents on mental health conditions have been published. 48 , 49 Even accounting for these added resources, since our initial review in 2020, the absolute number of these mental health–related publications among the total number of ACOG clinical guidance documents and among the total pool of OBGYN resident study materials remains low. There are likely other updated materials from other reviewed resources that have been published since our initial review, which is a limitation of the current study, although our results remain valid as a cross-sectional baseline for this novel analysis. 18 Reviews similar to this one may allow for examination of trends in this content area over time, and our hope is that bringing attention to this gap may lead to an increase of mental health topics incorporated into these resources in future editions. This study highlights the mismatch between the mental health content contained within these resources and the frequency with which OBGYN specialists interface with and provide clinical care for patients with these conditions on a regular basis. The prevalence of any mental illness among U.S. women is 27.2% (7% with serious mental illness), with a particular burden among those of reproductive age and rates increasing to more than 50% among those with additional socioeconomic risk factors. 50 The COVID-19 pandemic magnified these concerns because it exacerbated known risk factors for mental health disorders, leading to increasing prevalence of anxiety and depression, with young women more severely affected. 51 Prevalence of anxiety and depression symptoms among this group has not return to prepandemic levels, suggesting that those who become pregnant in the postpandemic era will have ongoing increased mental health care needs in the perinatal period. 52 Perinatal mental health conditions are one of the most common complications of pregnancy and childbirth, more common than diabetes, 53 hypertensive disorders of pregnancy, 54 and preterm birth, 55 and are among the leading causes of maternal mortality in the United States. 56 Furthermore, for many, a yearly gynecologic annual exam or routine prenatal care visits may be the only interactions a patient has with a primary care practitioner when mental health needs may be identified or addressed. Given that these conditions are highly prevalent, addressing care of basic mental health conditions is of significant relevance for OBGYN specialists and residency education. This review highlights the current noticeable lack of content related to commonly encountered mental health topics other than depression and selective serotonin reuptake inhibitor prescribing, and results may be used to organize efforts to incrementally expand materials prepared for, and used by, OBGYN residents to study and learn during residency. For example, the U.S. Preventive Services Task Force and ACOG both recommend screening for perinatal anxiety, and the CREOG Educational Objectives for OBGYN residents suggests there should be curriculum dedicated to posttraumatic stress disorder, although these topics were infrequently included in study materials. More complex psychiatric conditions, such as obsessive-compulsive disorder, bipolar disorder, and schizophrenia, were rarely covered, although prevalence of these conditions in the perinatal period is reported to be as high as 20% in some studies. 57 , 58 There are OBGYN-specific considerations related to psychopharmacologic medication use in pregnancy and lactation and notable drug interactions with pharmacologic contraceptives and treatments for endometriosis and other common gynecologic conditions that are not contained in these resources. Among many possible opportunities for expanding mental health–related content exposure for OBGYN residents, increasing the depth and breadth of what is contained within available study materials for residents is an important contributor. As such, program directors for OBGYN residencies preparing yearly education curricula can use the findings of this study to understand where gaps in resident education and study materials may be supplemented by local expert lecturers in maternal mental health or relevant reading materials. Other proposed solutions to fill these educational gaps are to introduce evidence-based curricula designed specifically to address mental health education among OBGYN residents. There have been a small number of successful pilot curriculum programs tested cross-sectionally among OBGYN residents at single centers, and there is ongoing development and multicenter efficacy testing underway for a national curriculum in reproductive psychiatry for OBGYN residents called NCRP × OB. 10 , 11 , 59 , 60 These programs endeavor to bundle highly salient and clinically useful educational learning points for baseline OBGYN resident knowledge into a series of lectures and educational activities that can be delivered easily and without significant resource strain during residency didactic lectures. These types of novel approaches, structures, and supports for teaching maternal mental health are viable strategies to accomplish the integration of these topics into OBGYN resident education going forward and can be made easily accessible and widely distributable through digital platforms. Finally, OBGYN learners particularly interested in reproductive psychiatry topics can be encouraged to seek additional exposure and mentorship through professional societies, interest groups, fellowships, conferences, and online toolkits. 61 Results of this study have some important limitations. It was not possible to review every independent third-party resource used by OBGYN residents for independent learning given time and resource constraints, although the inclusion criteria endeavored to contain an evidence-based, diverse, and balanced group of these materials. 17 , 18 Materials were reviewed in their most recent form in July 2020, which means that some editions have been updated and additional materials published or added to the question bank in the interim period. Although we report cumulative content contained in these resources, OBGYN residents may use none, one, or various combinations of these resources in independent learning, resulting in variable exposure to the content and therefore limitations in generalizability of the findings. We were also unable to review institution-specific OBGYN residency program educational resources (i.e., faculty lecture series, rotations in reproductive mental health clinics) for their contribution to mental health–related resident education. Finally, although this study reports on findings that suggest little current material is available for OBGYN residents to learn core mental health topics and although OBGYN specialist familiarity with psychiatric topics is necessary based on the prevalence of these conditions among pregnant and reproductive aged women and concurrent unmet needs identified by research publications, there should always be a collaboration among OBGYN specialists, psychiatrists, and mental health professionals to provide comprehensive mental health care to women. In conclusion, despite the critical importance and relevance of mental health care in comprehensive OBGYN care, this study demonstrated that mood and anxiety disorder content is underrepresented in commonly used OBGYN residency educational study resources, and existing resources do not comprehensively address all relevant aspects of these topics. These findings underscore an urgent need for more comprehensive inclusion of mental health topics in OBGYN residency study materials and resident education to better prepare graduates for addressing the complex mental health needs of their patients and ensure well-rounded, patient-centered care.

Supplemental digital content for this article is available at http://links.lww.com/ACADMED/B663 .