Objective
The purpose of this study was to estimate the frequency of identification of major depressive disorder by providers during prenatal care.
Study Design
A cohort of pregnant women who were participating in a randomized controlled trial and who had received a diagnosis of major depressive disorder was examined. Women were included in the current study if prenatal clinic records were available and legible.
Results
Clinical depression was noted in 56% of prenatal charts and on 24% of problem lists. Physicians and certified nurse midwives noted depression equally ( P = .935); physicians more frequently noted mental health referral (23% vs 0%; P = .01), and midwives more frequently included depression on the problem list ( P = .01). Recent medication use, which was stopped before conception or study participation, predicted notation of depression in the chart ( P = .001).
Conclusion
Depression frequently is missed during pregnancy and, when identified, is underacknowledged as a problem. Women who have not recently used antidepressant medication are more likely to be missed. Better screening and acknowledgment are needed.
Women are twice as likely as men to experience major depressive disorder (MDD) during their lives; the peak of first onset occurs during the childbearing years, between the ages 25 and 44 years. An estimated 7.1% of women will experience a new episode of MDD during pregnancy. From conception to birth, the estimated period prevalence of MDD is 12.7%; from birth to 2 months after delivery, the estimated period prevalence is 8.7%. Estimates are higher when both MDD and minor depression are considered.
The perinatal consequences of untreated depression and depressive symptoms alone are significant. Untreated maternal depression during pregnancy can result in serious maternal consequences such as suicide and in more severe postpartum depression and a multitude of fetal, neonatal, and childhood problems that include increased fetal heart rate, altered fetal habituation to external stimuli, changes in fetal DNA methylation patterns, placental gene expression, newborn infant inconsolability and excessive crying, and childhood psychopathologic conditions. Among low-income women, depressive symptoms alone have also been associated with an increase in intrauterine growth restriction, preterm birth, poor maternal weight gain, and alcohol, cigarette, and cocaine use.
Identification of depression during pregnancy is challenging and is hindered by the substantial overlap between the somatic symptoms of depression and normal symptoms of pregnancy (such as low energy, somatization, appetite changes, and disturbed sleep). Depressive symptoms are often overlooked during pregnancy, and most of the women who screen positive remain untreated. The degree to which MDD is recognized during pregnancy is largely unknown. Our objective was to estimate the identification of MDD by prenatal providers among a unique cohort of pregnant women who had been diagnosed independently with MDD.
Methods
A cohort of pregnant women who met the Diagnostic and Statistical Manual for Mental Disorders (4th edition) criteria for MDD and who scored at least 14 on the 17-item Hamilton Rating Scale for Depression was examined. The cohort consisted of women who originally were recruited to a randomized, blinded, controlled trial (RCT) that was designed to test acupuncture vs massage for depression during pregnancy. Participants were recruited between 2003 and 2008 through advertising in parent and baby publications and from local clinics.
Enrollment criteria for the original RCT were age ≥18 years, gestational age between 12 and 30 weeks, MDD diagnosis based on the Diagnostic and Statistical Manual for Mental Disorders (4th edition), text revision, and a score of ≥14 on the 17-item Hamilton Rating Scale for Depression.
Trained research assistants conducted the Structured Clinical Interview for Diagnostic and Statistical Manual for Mental Disorders (4th edition), Axis I to diagnose MDD and the Hamilton Rating Scale for Depression interview to document depressive symptom severity. Exclusion criteria from the original study were any of the following: other primary Axis I disorder in the past 2 months (except social phobia), seasonal affective disorder, psychotic features, abnormal thyroid panel, positive drug screen results (illicit drugs, hypnotics, or opiates), serious uncontrolled medical conditions or conditions that may be a medical basis of depression, cluster B personality disorders, current psychotherapy, herbs with psychotropic effects, psychotropic medications (eg, antidepressants), electroconvulsive therapy or vagal nerve stimulation in the last year, current active suicidal potential that necessitated immediate treatment, absence of prenatal care, and conditions for which bed rest was likely to be prescribed (eg, history of early preterm birth, poorly controlled hypertension). Women were recruited to the RCT between 2003 and 2008 by community and clinic-based advertising; all of the women were self-referred. Women in the original study received prenatal care from multiple providers throughout the greater San Francisco Bay Area. Authorizations to release prenatal records to the study were signed by all patients in the current study. In cases in which prenatal records were not received, multiple follow-up phone calls and letters were sent by study coordinators for up to 2 years after the study closed; updated consents were sought as needed.
The study was approved by the Human Subjects Committees at Stanford University Medical Center. All women signed study consent. As a part of consent for the study, women were advised about standard treatments of depression (psychotherapy and antidepressant medications). The consent informed women that, during their participation in the study of acupuncture vs massage for treatment of depression, they could not receive nonstudy treatment for depression and that at any time they were free to discontinue participation in the study with no ill effects. All women who participated in the RCT were encouraged to inform their prenatal care provider about their participation in the study. They were provided a form letter to their provider that stated “I am writing to let you know that I have enrolled in a study of depression during pregnancy at Stanford University Medical School. If I qualify, I will be randomized to receive acupuncture or massage. I have attached the study brochure and a letter that the study coordinator asked me to forward to you.” Study participants were encouraged to give the letter to their provider; the study did not keep track of whether the letter was sent. The study coordinator did not contact pregnancy care providers regarding study participation.
All study participants underwent an extensive phone screen by trained research assistants; the screening included demographic information, pregnancy history, depression history and treatments, attitudes toward antidepressant treatments, and in whom the patient confided her depression.
Women were included in the current cohort study if their prenatal clinic records were available and legible. Prenatal records were assessed for any acknowledgement by any provider of depression at the prenatal intake visit, prenatal clinic visits, and the antepartum chart problem list. Depression was considered as having been noted in the record if the following words and notations were recorded anywhere in the record: depression , major depressive disorder , Stanford Depression Study , reference to the patient’s emotional state that included “mood,” “stress,” “anxiety,” “doing well emotionally,” or “feels good,” or reference to follow up with a psychiatric professional. Although the latter words may not represent acknowledgment of depression in every case, given the limits of a chart review and potential for variability in how a caregiver might describe depression, we chose a conservative approach to our definitions to avoid overstatement of the primary outcome. All records were assessed by a single reviewer (D.J.L.).
All data were entered into a database (Access software; Microsoft Corporation, Redmond, WA). Statistical tests included the χ 2 test of proportions, Fisher exact tests, and logistic binary regression and were considered significant with a probability value of < .05.
Results
Of 134 women in the parent study who completed at least 1 follow-up session after random assignment, complete prenatal records that were available for 89 women were analyzed in the current study. Among those women who were not included, records for 24 women were incomplete (did not include the prenatal intake visit, prenatal clinic visits, and the antepartum chart problem list); the medical record release form for 17 women was not signed or had expired, and requested records for 3 women were not sent by providers. One record was excluded because of illegible handwriting. Demographic, obstetric, and depression history characteristics are shown in Table 1 . Seventy-five women were cared for primarily by obstetricians; 9 women were cared for by Certified Nurse-Midwives; 2 women were cared for by lay midwives, and the provider type could not be determined from the charts of 3 women.
| Characteristic | Measure |
|---|---|
| Maternal age, y a | 33.31 ± 4.70 |
| Gestational age at study entry, wk a | 20.5 ± 5.75 |
| Parity, n a | 0.65 ± 0.92 |
| Race/ethnicity, n (%) | |
| Black | 2 (2) |
| Asian | 15 (17) |
| White | 39 (44) |
| Other | 33 (37) |
| History of previous depression, n (%) | 79 (89) |
| Hamilton Rating Scale for Depression score at study entry a | 20.4 ± 3.7 |
| Onset of depression before pregnancy, n (%) | 38 (43) |
| Onset of depression during pregnancy, n (%) | 51 (57) |
| Taking medications before pregnancy, n (%) | 34 (38) |
| Seeing a therapist at study entry, n (%) | 10 (8) |
| Discontinuation of medications before conception, n (%) | 7 (8) |
| Discontinuation of medications after conception, n (%) | 17 (9) |
Among the entire cohort, depression was noted in the prenatal charts for 56% of the women and was never noted for 44% ( Table 2 ). When depression was noted in the chart, it was noted at the intake visit for 39% of the women, at any subsequent clinic visit for 37% of the women, and on the problem list for 24% of the women. Among the 35 participants whose depression was noted at the intake visit, depression was listed on the problem list for 21 women (60%). Depression was noted once only for 20 women (at the intake visit only and not anywhere else in the chart, among 7 women [8%]), twice for 12 women (13%), 3 times for 15 women (17%), 4 times for 5 women (6%), and at all visits for 1 woman (1%). Participation in the Stanford RCT was mentioned in 22 charts (25%). Certified Nurse-Midwives were more likely to note depression as a problem on the problem list, and physicians were more likely to note that a referral to mental health consultation had been made ( Table 3 ). There were no significant differences in acknowledgment of depression in the chart between physicians and Certified Nurse-Midwives (68% vs 67%; P = .935).
| Variable | Depression acknowledged on chart, n (%) |
|---|---|
| Depression noted in prenatal chart | 50 (56) |
| Depression never noted in the chart | 39 (44) |
| Intake visit | 35 (39) |
| Problem list | 21 (24) |
| Clinic visits | 33 (37) |
| Delivery hospitalization | 14 (16) |
| Postpartum visit | 11 (12) |
| Intake visit, only | 7 (8) |
| Problem list, only | 1 (1) |
| Clinic visit, only | 9 (10) |
| Delivery hospitalization, only | 1 (1) |
| Postpartum visit, only | 2 (2) |
| Problem list if noted at intake visit | 21 (24) |
| First noted during delivery hospitalization | 1 (1) |
| First noted postpartum | 2 (2) |
| Variable | Doctor (n = 75) | Certified nurse-midwife (n = 9) | P value |
|---|---|---|---|
| Depression on problem list | 14 (19%) | 5 (56%) | .01 |
| Professional referral made | 17 (23%) | 0 | .01 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
