Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 3.
Published in final edited form as: J Affect Disord. 2022 Jan 15;303:82–90. doi: 10.1016/j.jad.2022.01.064

A comparison of symptoms of bipolar and unipolar depression in postpartum women

Crystal T Clark 1, Dorothy K Sit 1, Katelyn B Zumpf 1, Jody D Ciolino 1, Amy Yang 1, Sheehan D Fisher 1, Katherine L Wisner 1,*
PMCID: PMC13044965  NIHMSID: NIHMS2158839  PMID: 35041868

Abstract

Background:

Distinguishing postpartum women with bipolar from unipolar depression remains challenging, particularly in obstetrical and primary care settings. The post-birth period carries the highest lifetime risk for the onset or recurrence of Bipolar Disorder (BD). Characterization of differences between unipolar and bipolar depression symptom presentation and severity is critical to differentiate the two disorders.

Methods:

We performed a secondary analysis of a study of 10,000 women screened by phone with the Edinburgh Postnatal Depression Scale at 4–6 weeks post-birth. Screen-positive mothers completed the Structured Clinical Interview for DSM-4 and those diagnosed with BD and unipolar Major Depressive Disorder (UD) were included. Depressive symptoms were assessed with the 29-item Structured Interview Guide for the Hamilton Rating Scale for Depression (SIGH-ADS).

Results:

The sample consisted of 728 women with UD and 272 women with BD. Women with BD had significantly elevated levels of depression severity due to the higher scores on 8 of the 29 SIGH-ADS symptoms. Compared to UD, women with BD had significantly higher rates of comorbid anxiety disorders and were twice as likely to report sexual and/or physical abuse.

Limitations:

Only women who screened positive for depression were included in this analysis. Postpartum women with unstable living situations, who were hospitalized or did not respond to contact attempts did not contribute data.

Conclusions:

Severity of specific symptom constellations may be a useful guide for interviewing postpartum depressed women along with the presence of anxiety disorder comorbidity and physical and/or sexual abuse.

Keywords: Bipolar disorder, Perinatal, Pregnancy, Postpartum, Unipolar, Screening

1. Introduction

Postpartum depression is a complication of childbirth that affects one in seven women (Gaynes et al., 2005; Wisner et al., 2013) and has adverse health consequences for the mother, her infant, and family (Field, 2010). Due to its high prevalence and adverse outcomes, depression screening is recommended by the US Preventive Services Task Force (Siu et al., 2016), the American Psychiatric Association (Byatt et al., 2018) and the American College of Obstetricians and Gynecologists (American College of Obstetricians and Gynecologists, 2018). The most widely used postpartum screening tool is the Edinburgh Postnatal Depression Scale (EPDS), a 10-item questionnaire available in more than 20 languages and validated in multiple racial and socioeconomic groups (Gibson et al., 2009; Cox and Holden, 2003). Although screening has increased the rate of identification of women with perinatal depression, the challenge of distinguishing unipolar major depressive (UD) from bipolar disorder (BD) remains substantial, particularly in obstetrical and primary care settings (Frye et al., 2005).

The difficulty differentiating depression in the context of UD from BD results in a nearly 10-year delay in the diagnosis of BD for many patients (Baethge et al., 2003; Fritz et al., 2017; Ghaemi et al., 2002). The post-birth period marks the highest risk for first onset or recurrence an episode consistent with BD in a woman’s lifetime (Kendell et al., 1987; Munk-Olsen et al., 2006). Although unfortunate for new mothers, this predictable time of high incidence affords a unique opportunity for accurate mood disorder diagnosis. One in four postpartum women with BD experience recurrent affective episodes. Within the first 30 days post-delivery, the hospitalization risk is significantly higher for women with BD (Relative Risk [RR] = 23.3, 95% CI=11.5–47.2) than for women with UD (RR=2.8, 95% CI=1.9–4.1) (Munk-Olsen et al., 2006). Postpartum psychotic episodes occur in 15–40% of women with BD (Blackmore et al., 2013) compared to 1.5% of women with UD (Di Florio et al., 2013). In a population-based Scandinavian study, 14% of women with first-time psychiatric contacts within 30 days after birth converted to a diagnosis of BD within 15 years compared to 4% of women with a first-time psychiatric contact unrelated to childbirth (Munk-Olsen et al., 2012). Psychiatric disorders contribute to the increasing rate of American maternal mortality in the first post-birth year. Overdose or poisoning and suicide were the second and seventh leading causes of death and comprised 8.8% of maternal deaths (Davis et al., 2019).

Tools to differentiate postpartum women with BD from UD are urgently needed. Accurate diagnosis and treatment are critical in this vulnerable population of new mothers with BD to avoid antidepressant monotherapy, which is associated with rapid-cycling and increased depressive morbidity (El-Mallakh et al., 2015). The Mood Disorder Questionnaire (MDQ) is a 13-item self-report checklist of symptoms of BD with two supplemental questions (Hirschfeld, 2002). The patient must recognize and report moderate impairment while experiencing at least seven of thirteen criteria for a positive screen. Masters et al. (2019) and Merrill et al. (2015) reported that 18.8% and 21.4%, respectively, of perinatal, non-psychiatric population screened positive on the MDQ. The addition of the MDQ to the EPDS improved the differentiation of UD from BD during screening for postpartum depression. Identification of women who had BD (established by the Structured Interview for DSM-4, SCID) was achieved for 50% of women with application of the original MDQ scoring criteria; however, nearly 70% of women with BD were identified when the MDQ was scored with omission of the impairment criterion (Clark et al., 2015). Although an improvement in case identification, 30% of women remained incorrectly diagnosed.

Differences in the intensity and type of depressive symptoms in patients with BD compared to UD are useful for diagnostic evaluations. Symptoms associated more frequently with BD than UD include agitation and atypical symptoms (hypersomnia, increased appetite and weight gain, oversensitivity to rejection, leaden paralysis, and diurnal variation with mood worsening later in the day) (Perugi et al., 1998). In a general population of patients with mood disorders who were included in clinical trials, Perlis et al. (2006) reported that patients with BD had higher scores on the Montgomery-Asberg Rating Scale depression score (30.9 ± 6.00) compared to those with UD (27.7 ± 5.7-study 1; 26.6 ± 5.8- study 2, p <0.001). The individual depressive symptoms that significantly differed between individuals with BD and UD were apparent sadness, tension, reduced sleep, pessimistic thoughts, and suicidal thoughts. Patients with BD have a higher risk for suicide attempts and completions compared to most psychiatric disorders, including UD (Ahrens et al., 1995).

Comorbid anxiety and substance use disorders also are more prevalent among general populations of patients with BD compared to those with UD (Merikangas et al., 2007). Postpartum women with BD (n = 30) experienced significantly more obsessive compulsive (p = 0.036) and posttraumatic stress disorder (p = 0.014) comorbidity compared to women with UD (n = 26) (Sharma et al., 2008). Perinatal women with BD are more likely to have substance use comorbidity compared to women with other diagnoses (Battle et al., 2014). In a naturalistic study of the illness course of pregnant women with BD (N = 152), women reported smoking cigarettes (40%) and illicit drug (27%) and alcohol (19%) use (Driscoll et al., 2017).

Accurate and timely diagnosis in the perinatal period is critical to provide evidence-based treatment and improve outcomes for the woman, her infant, and family. Characterization of the constellation of depressive symptoms specifically in postpartum women with BD compared to UD has received minimal research attention. We compared the symptom sets, psychiatric comorbidities, and functional levels of postpartum women with BD and UD. We hypothesized that postpartum women with BD would have significantly greater frequencies of: 1) agitation, 2) suicidal ideation, 3) atypical depressive symptoms, 4) anxiety disorder comorbidity, 5) past trauma, and 6) impaired function.

2. Methods

The primary screening study of 10,000 postpartum women has been described in detail (Wisner et al., 2013) and is briefly reviewed here. The screening program was conducted at a large urban Midwest maternity hospital. Women who delivered were visited by a nurse or social worker on the maternity unit and offered phone screenings with the EPDS (Cox et al., 1987) at 4–6 weeks post-birth. Women who were non-English speaking, younger than 18 years, unable to provide consent, or did not have a telephone available for follow-up contact were excluded. Eligible women signed a waiver approved by the Institutional Review Board, which allowed collection of contact information and telephone screening data. The 4to 6-week period after birth was chosen because it covers the peak period of risk for outpatient psychiatric contacts, which is the highest through the first 19 days and persists for the first 3 post-partum months (Munk-Olsen et al., 2006).

An intense effort, with day, evening and weekend calls, was made to reach the participants who agreed to be contacted. If a participant was not reached within 3 business days, a postcard encouraging the mother to contact the team was sent. Women who were not reached by week 6 were removed from the contact list and no further contact was attempted. Women who were contacted and had positive EPDS scores (defined by a score of ≥10) were offered home visits and psychiatric assessments within 2 weeks of the screen.

All mothers who agreed to a home visit completed the SCID during the visit. To maximize inclusivity and respect women’s privacy in the early post-birth period, women who declined the home visit but were willing to participate were interviewed by telephone. The SCID established primary and secondary (comorbid) psychiatric diagnoses. For the purposes of this study, the primary SCID diagnoses were grouped into three categories (UD, BD and other disorder). The severity of women’s depressive symptoms was assessed with the 29-item Structured Interview Guide for the Hamilton Rating Scale for Depression, Atypical Depression Symptom Version (SIGH-ADS) (Williams and Terman, 2003), which includes the 17 and 21-item versions of the Hamilton Depression Rating Scale (Hamilton, 1960) and an eight-item supplemental scale for atypical depressive symptoms. Each SIGH-ADS symptom is scored on a severity scale with 2, 3, or 4 points. Postpartum function and support were assessed with the Short Form Health Survey-12 (SF-12) (Ware et al., 1996). Four questions related to physical and sexual abuse experienced as a child or adult were presented. We summed the number of positive responses to create a variable from 0 to 4 for analytic purposes.

2.1. Statistical analysis

We used descriptive statistics to summarize participant demographics, comorbidities, and symptomatology. Frequencies and percentages were recorded for all categorical variables and mean (with standard deviation) for numeric variables. To determine whether an association existed between participant demographics, comorbidities, or symptomatology and diagnosis (BD vs. UD), we performed either a Pearson’s Chi-Squared test, Fisher’s Exact test (in the case of low cell counts for categorical variables), or Student’s t-test. We corrected for multiple hypothesis testing using the Holm’s procedure (Holm, 1979), and included all statistical hypothesis tests conducted in the procedure.

As a sensitivity analysis, we performed multiple variable linear regression for the following outcomes: SIGH-ADS 29, 21, or 8-item (atypical depressive) total symptom scores. We adjusted for demographics and comorbidities that correlated with these outcomes at a value of 0.3 or greater regardless of statistical significance. We summed the scores of the eight symptoms of the SIGH-ADS 29 that were associated significantly with diagnosis and developed a receiver-operating characteristic (ROC) curve. Our goal was to explore whether the eight symptoms most associated with BD in this sample had any clinical utility to differentiate BD from UD in postpartum women. To assess the performance of the derived score the area under the curve (AUC) was calculated in addition to sensitivity, specificity, positive predicted value, and negative predicted value at various cut-off scores. The optimal cut-off point was determined by maximizing the sum of sensitivity and specificity. All analyses were conducted using R (version 3.5.3, 2019, The R Foundation) and assumed a two-sided, 5% level of significance.

3. Results

3.1. Sample characteristics

The derivation of the study sample is detailed in Fig. 1. A total of 15,172 women were screened with the EPDS and 2019 (13%) screened positive. All screen-positive women were asked to complete the SCID and SIGH-ADS interviews and 1190 accepted the diagnostic interviews either in-person at home or by telephone. Details on the demographics of women who accepted home visits versus telephone visits in a subset of this population have been previously published (Wisner et al., 2013). Women who accepted home visits versus telephone assessments or who declined interviews were more likely to have higher mean EPDS scores (14.3 [3.9] vs 12.3 [3.0] vs 13.3 [3.7], respectively), to be Black, publicly insured or uninsured, single, younger, and less educated.

Fig. 1.

Fig. 1.

Open in a new tab

Study flow chart.

Women with primary diagnoses of UD (n = 728) and BD (n = 272) were included in these analyses. The BD group consisted of all subtypes including BD I (n = 132), BD II (n = 86), and BD NOS (n = 54). We highlight the major findings below; however, we refer the reader to the Tables for all details of analysis results, including results of individual statistical tests.

The demographic characteristics of the women grouped by diagnosis are displayed in Table 1. Women with UD were older, more educated, and more likely to be married than those with BD. The SF12 Mental Scale (BD: 30.7 ± 10.4 vs. UD: 31.4 ± 9.7) and the SF12 Physical scale (BD: 47.6 ± 10.2 vs. UD: 49.2 ± 9.5) did not significantly differ between women with BD compared to UD. The experience of sexual and physical abuse as a child or adult was common in this group of women with mood disorders. All types of abuse were about twice as common as women with BD compared to those with UD. Women who were diagnosed with BD experienced childhood physical (40.6%) and sexual (43.8%) abuse. In adulthood, 53% of subjects reported physical and 27% sexual abuse.

Table 1.

Participant characteristics.

Level Diagnosis Analysis
BD UD p-Value Holm’s Adjusted p-value
N = 272 N = 728
Age (mean (SD)) 26.22 (5.27) 28.33 (5.88) <0.0001 0.0054
Race (%) Black 85 (31.2) 177 (24.3) 0.011 0.297
Other 24 (8.8) 43 (5.9)
White 163 (59.9) 508 (69.8)
Hispanic (%) 13 (5.0) 14 (2.0) 0.0273 0.6006
Education (%) <High School 46 (16.9) 54 (7.4) <0.0001 0.0054
High school 83 (30.5) 165 (22.7)
Some college 106 (39.0) 239 (32.8)
College 22 (8.1) 151 (20.7)
Graduate school 15 (5.5) 119 (16.3)
Marital Status (%) Divorced/Widowed 12 (4.4) 17 (2.3) <0.0001 0.0054
Married 67 (24.6) 371 (51.0)
Single 193 (71.0) 340 (46.7)
Parity (%) 1 97 (35.7) 261 (35.9) 0.0023 0.0713
2 77 (28.3) 283 (38.9)
3 60 (22.1) 118 (16.2)
4+ 38 (14.0) 66 (9.1)
SF12 Mental Scale (mean (SD)) 30.7 (10.4) 31.4 (9.7) 0.4639 >0.9999
SF12 Physical Scale (mean (SD)) 47.6 (10.2) 49.23 (9.5) 0.0896 >0.9999
Physically abused as a child (%) Yes 104 (40.6) 139 (19.6) <0.0001 0.0054
No 152 (59.4) 570 (80.4)
Sexually abused as a child (%) Yes 112 (43.8) 174 (24.5) <0.0001 0.0054
No 144 (56.2) 535 (75.5)
Physically abused as an adult (%) Yes 136 (53.1) 238 (33.6) <0.0001 0.0054
No 120 (46.9) 471 (66.4)
Sexually abused as an adult (%) Yes 69 (27.0) 99 (14.0) <0.0001 0.0054
No 187 (73.0) 610 (86.0)
Trauma (%) * 0 73 (26.8) 347 (47.7) <0.0001 0.0054
1 65 (23.9) 193 (26.5)
2 69 (25.4) 124 (17.0)
3 40 (14.7) 47 (6.5)
4 25 (9.2) 17 (2.3)
Open in a new tab
*

Trauma= the total of the types of abuse above (0–4).

3.2. Comorbid diagnoses

(Table 2). With the exception of GAD, women with BD were more likely to report a comorbid anxiety disorder (as determined by DSM 4 criterion) compared to women with UD. Panic disorder (27.2% vs.12.0%), post-traumatic stress disorder (22.8% vs. 9.3%) and obsessive-compulsive disorder (21.3% vs.10.7%) were more prevalent among women with BD compared to UD. Substance use disorders were more common in women with BD and eating disorders were more common in women with UD, but these differences were not statistically significant.

Table 2.

Comorbid diagnoses.

Diagnosis Analysis
BD UD p-Value Holm’s Adjusted p-value
N = 272 N = 728
GAD and Anxiety Disorder NOS (%) 112 (41.2) 327 (44.9) 0.3226 >0.9999
Panic Disorder: (%) 74 (27.2) 87 (12.0) <0.0001 0.0054
Obsessive-Compulsive Disorder (%) 58 (21.3) 78 (10.7) <0.0001 0.0054
Post-traumatic Stress Disorder (%) 62 (22.8) 68 (9.3) <0.0001 0.0054
Substance Use (%) 33 (12.1) 59 (8.1) 0.066 >0.9999
Eating Disorder (%) 8 (2.9) 28 (3.8) 0.6221 >0.9999
Open in a new tab

3.3. Depressive symptoms

Compared to women with UD, women with BD had significantly higher levels of depression severity on the SIGH-ADS-29 (UD: 21.0 ± 5.7 vs. BD: 23.9 ± 6.7), respectively, unadjusted p-value<0.0001; Holms adjusted p-value = 0.0054) and the SIGH-ADS-21 (UD: 15.7 ± 4.6 vs. BD: 18.1 ± 5.4, respectively, unadjusted p-value<0.0001, Holms adjusted p-value= 0.0054) (Table 3). However, contrary to our hypothesis, the SIGH-ADS Atypical Depression subscale did not significantly differ between women with UD and BD (UD 5.3 ± 2.4 vs. BD 5.7 ± 2.8, unadjusted p-value=0.0123, Holms adjusted p-value=0.3198) (Table 3).

Table 3.

Symptomatology.

Diagnosis Analysis
Score BD N = 272 UD N = 728 p-Value Holm’s Adjusted p-value Test
SIGH-ADS 29 Item Total (mean (SD)) 23.85 (6.69) 20.97 (5.66) <0.0001 0.0054
HDRS-21 Item Total (mean (SD)) 18.11 (5.42) 15.68 (4.57) <0.0001 0.0054
Depressed Mood (%) 0 9 (3.4) 35 (4.8) <0.0001 0.0054 exact
1 100 (37.5) 392 (53.8)
2 157 (58.8) 295 (40.5)
3 1 (0.4) 5 (0.7)
4 0 (0.0) 1 (0.1)
Work and Activities (%) 0 30 (11.2) 70 (9.6) 0.0076 0.2204 exact
1 72 (27.0) 281 (38.6)
2 160 (59.9) 359 (49.3)
3 5 (1.9) 17 (2.3)
4 0 (0.0) 1 (0.1)
Genital Symptoms (%) 0 84 (31.5) 214 (29.4) 0.1211 >0.9999
1 60 (22.5) 211 (29.0)
2 123 (46.1) 303 (41.6)
Somatic Symptoms Gastrointestinal (%) 0 78 (29.2) 271 (37.2) <0.0001 0.0054
1 123 (46.1) 394 (54.1)
2 66 (24.7) 63 (8.7)
Loss of Weight: History (%) 0 186 (71.3) 505 (70.1) 0.9498 >0.9999 exact
1 66 (25.3) 190 (26.4)
2 9 (3.4) 25 (3.5)
Loss of Weight: Actual (%) 0 1 (20.0) 2 (25.0) 0.4615 >0.9999 exact
1 0 (0.0) 3 (37.5)
2 4 (80.0) 3 (37.5)
Carbohydrate Craving or Eating since pregnancy or last visit (%) 0 101 (37.8) 317 (43.6) 0.1181 >0.9999
1 166 (62.2) 410 (56.4)
Insomnia Early (Sleep Onset Insomnia) (%) 0 107 (40.1) 412 (56.6) <0.0001 0.0054
1 42 (15.7) 142 (19.5)
2 118 (44.2) 174 (23.9)
Insomnia Middle (Sleep Maintenance Insomnia) (%) 0 46 (17.2) 98 (13.5) 0.0278 0.6006
1 58 (21.7) 217 (29.8)
2 163 (61.0) 413 (56.7)
Insomnia Late (Early Awakening) (%) 0 150 (56.2) 438 (60.2) 0.0191 0.4775
1 55 (20.6) 176 (24.2)
2 62 (23.2) 114 (15.7)
Somatic Symptoms General (%) 0 39 (14.6) 102 (14.0) 0.0422 0.8018
1 123 (46.1) 397 (54.5)
2 105 (39.3) 229 (31.5)
Feelings of Guilt (%) 0 45 (16.9) 105 (14.4) <0.0001 0.0054 exact
1 95 (35.6) 393 (54.0)
2 126 (47.2) 226 (31.0)
3 1 (0.4) 3 (0.4)
4 0 (0.0) 1 (0.1)
Suicide (%) 0 213 (79.8) 668 (91.8) <0.0001 0.0054 exact
1 43 (16.1) 46 (6.3)
2 9 (3.4) 11 (1.5)
3 1 (0.4) 3 (0.4)
4 1 (0.4) 0 (0.0)
Anxiety Psychic (%) 0 13 (4.9) 35 (4.8) 0.068 >0.9999 exact
1 58 (21.7) 213 (29.3)
2 185 (69.3) 463 (63.6)
3 11 (4.1) 16 (2.2)
4 0 (0.0) 1 (0.1)
Anxiety Somatic (%) 0 35 (13.1) 88 (12.1) 0.0952 >0.9999 exact
1 96 (36.0) 324 (44.5)
2 127 (47.6) 298 (40.9)
3 9 (3.4) 18 (2.5)
Hypochondriasis (%) 0 156 (58.4) 429 (58.9) 0.2331 >0.9999 exact
1 82 (30.7) 242 (33.2)
2 27 (10.1) 56 (7.7)
3 1 (0.4) 1 (0.1)
4 1 (0.4) 0 (0.0)
Insight (%) 0 260 (97.4) 682 (93.7) 0.005 0.15 exact
1 6 (2.2) 46 (6.3)
2 1 (0.4) 0 (0.0)
Retardation (%) 0 184 (68.9) 555 (76.2) 0.0362 0.724 exact
1 75 (28.1) 162 (22.3)
2 8 (3.0) 11 (1.5)
Agitation (%) 0 158 (59.2) 589 (80.9) <0.0001 0.0054 exact
1 69 (25.8) 110 (15.1)
2 25 (9.4) 23 (3.2)
3 14 (5.2) 5 (0.7)
4 1 (0.4) 1 (0.1)
Severity of Variation (%) 0 130 (48.7) 283 (38.9) 0.0191 0.4775
1 76 (28.5) 256 (35.2)
2 61 (22.8) 189 (26.0)
Depersonalization and Derealization (%) 0 170 (63.7) 538 (73.9) 0.0018 0.0576 exact
1 78 (29.2) 166 (22.8)
2 19 (7.1) 22 (3.0)
3 0 (0.0) 2 (0.3)
Paranoid Symptoms (%) 0 181 (67.8) 619 (85.0) <0.0001 0.0054 exact
1 82 (30.7) 108 (14.8)
2 3 (1.1) 1 (0.1)
3 1 (0.4) 0 (0.0)
Obsessional and Compulsive Symptoms (%) 0 173 (64.8) 586 (80.5) <0.0001 0.0054
1 78 (29.2) 121 (16.6)
2 16 (6.0) 21 (2.9)
SIGH-ADS Atypical Item Total [mean (SD)] 5.74 (2.75) 5.29 (2.43) 0.0123 0.3198
Social Withdrawal (%) 0 53 (19.9) 187 (25.7) 0.0243 0.5589 exact
1 63 (23.6) 207 (28.4)
2 146 (54.7) 325 (44.6)
3 5 (1.9) 9 (1.2)
Weight Gain (%) 0 239 (89.5) 666 (91.5) 0.0903 >0.9999 exact
1 25 (9.4) 61 (8.4)
2 3 (1.1) 1 (0.1)
Appetite Increase (%) 0 227 (85.0) 623 (85.6) 0.646 >0.9999 exact
1 18 (6.7) 59 (8.1)
2 15 (5.6) 31 (4.3)
3 7 (2.6) 15 (2.1)
Increased Eating (%) 0 215 (80.5) 592 (81.3) 0.2367 >0.9999 exact
1 28 (10.5) 94 (12.9)
2 19 (7.1) 32 (4.4)
3 5 (1.9) 10 (1.4)
Carbohydrate Craving or Eating (%) 0 101 (37.8) 317 (43.5) 0.3839 >0.9999 exact
1 117 (43.8) 284 (39.0)
2 44 (16.5) 110 (15.1)
3 5 (1.9) 17 (2.3)
Hypersomnia (%) 0 244 (91.4) 683 (93.8) 0.1302 >0.9999 exact
1 5 (1.9) 15 (2.1)
2 11 (4.1) 25 (3.4)
3 3 (1.1) 3 (0.4)
4 4 (1.5) 2 (0.3)
Fatigability (%) 0 20 (7.5) 26 (3.6) 0.0082 0.2296 exact
1 67 (25.1) 211 (29.0)
2 113 (42.3) 293 (40.2)
3 60 (22.5) 150 (20.6)
4 7 (2.6) 48 (6.6)
Diurnal Variation Type B (%) 0 119 (44.6) 381 (52.3) 0.0606 >0.9999 exact
1 96 (36.0) 232 (31.9)
2 44 (16.5) 106 (14.6)
3 8 (3.0) 9 (1.2)
Open in a new tab

As a sensitivity analysis, we adjusted for SF12 Mental Scale. The results were consistent with the unadjusted results above with the estimated difference in SIGH-ADS-29 item of −2.25 (UD-BD β: 2.25, 95% CI: (−3.34, −1.16), p-value=0.0001), difference in HDRS-21 of −2.07 (UD-BD β: −2.07, 95% CI: ( −2.94, −1.20), p-value<0.0001), and difference in SIGH-ADS Atypical Depression subscale of −0.17 (UD-BD β: −0.17, 95% CI (−0.66, 0.32), p-value=0.4950) after controlling for SF12 Mental Scale.

We explored whether individual items on the SIGH-ADS were associated with diagnosis. Eight of the 29 symptoms assessed on the SIGH-ADS were associated with diagnosis and more frequent in women with BD compared to UD: depressed mood, somatic symptoms gastrointestinal, insomnia early, feelings of guilt, suicide, agitation, paranoid symptoms and obsessional and compulsive symptoms. We developed a sub-score of the SIGH-ADS 29 consisting of the sum of symptom severity ratings for the eight symptoms. The average composite score was 5.2, with median 5, interquartile range from 3 to 7, and ranged from 0 to 15. The receiver-operator characteristics (ROC) area under the curve (AUC) was 0.70 (0.67, 0.74) (Fig. 2). A composite score of 7 or more was the most favorable cut-point in identifying BD in this sample, with a sensitivity (the proportion of women that test positive for BD given they have BD) of 0.52 (0.46, 0.59) and specificity (the proportion of women that test negative for BD given they do not have BD) of 0.82 (0.79, 0.85) (Table 4). Assuming the prevalence of BD among those who screen positive for depression with the EPDS is 18.5%, (Masters et al., 2019) at a cut off of 7 or more, the positive predicted value (the probability of women that have BD given they test positive for BD) was 39.6%, and the negative predicted value (the probability of women that have MD given they test negative for BD) was 88.3%.

Fig. 2.

Fig. 2.

Open in a new tab

Receiver-operator characteristic curve.

Table 4.

Sensitivity/specificity of the 8-symptoms.

Sum of 8 Symptoms Cases Identified Sensitivity Specificity PPV* NPV*
>0 267 100% 0.8% 18.6% 100%
>1 260 97.4% 4.4% 18.8% 88.1%
>2 249 93.3% 13.9% 19.7% 90.1%
>3 229 85.8% 32% 22.3% 90.8%
>4 200 74.9% 49.6% 25.2% 89.7%
>5 171 64% 67.9% 31.1% 89.3%
>6 140 52.4% 81.9% 39.6% 88.3%
>7 92 34.5% 90.7% 45.6% 85.9%
>8 57 21.3% 95.5% 51.7% 84.2%
>9 35 13.1% 98.2% 62.5% 83.3%
>10 18 6.7% 99% 61.4% 82.4%
>11 9 3.4% 100% 100% 82.0%
>12 5 1.9% 100% 100% 81.8%
>13 1 0.4% 100% 100% 81.6%
>14 1 0.4% 100% 100% 81.6%
Open in a new tab
*

These data are based on an 18.5% of BD in a postpartum population (Masters et al., 2019).

4. Discussion

We observed higher mean depression scores in women with BD than UD (HRSD-21: BD=18.11 ± 5.42; UD=15.68 ± 4.57, p = 0.005). Although the symptoms of depression are generally similar between BD and UD, (Brockington et al., 1982; Endicott et al., 1985; Mitchell et al., 1992, 2001; Parker et al., 2000) our novel finding is the identification of eight symptoms that were significantly more frequent and severe in a large sample of newly postpartum women with BD compared to UD. Moses-Kolko et al. (2012) also reported higher mean symptom scores in postpartum women with BD versus UD on the 21-item HRSD (24.4 ± 4.0 and 19.9 ± 15.6, respectively). In the study by Perlis et al. (2006), two of the symptoms that were significantly more common in patients with BD than UD (suicidal thoughts and inner tension) parallel our findings of suicidality and agitation in postpartum women. However, contrary to our hypothesis, atypical symptoms were no more common in post-partum women with BD than UD.

In our exploratory ROC analysis, we found that a composite symptom score of 7 or more was the optimal cut-point in identifying post-birth BD with a marginal sensitivity of 0.52 and specificity of 0.82 for postpartum women. These values are not adequate for use as a screening tool. The developers of other screening measures for BD have incorporated historical and clinical variables in addition to symptoms to improve the capacity to differentiate BD from UD. For example, the MDQ includes a question on family history of bipolar disorder and Perlis et al. (2006) included early age of mood disorder onset, family history and number of depressive episodes. In non-postpartum women who were diagnosed with UD (n = 54) and BD (n= 28), the WHIPLASHED (Mahmoud et al., 2019) mnemonic had reasonable predictive capacity (AUC = 0.877) to differentiate BD from UD (89% sensitivity and 76% specificity for a diagnosis of BD). This interview includes clinical factors including historical information (W=worse taking antidepressants, S=seasonal or postpartum pattern, A=abrupt onset or termination of depressive episodes), symptoms (H=hypomania/hyperthymia, I=irritable, P=psychotic symptoms) as well as loaded family history (L). Until biomarkers are available to reliably differentiate BD from UD, detailed assessment of symptoms, comorbidities and historical characteristics will remain our primary diagnostic tools.

Our evaluation of factors other than symptoms included psychiatric comorbidities. In our sample, the ratio of patients with BD compared to UD was 2.5 for post-traumatic stress disorder, 2.0 for obsessive-compulsive disorder and 2.3 for panic disorder. A study of patients with BD who were misdiagnosed with UD demonstrated greater rates of generalized anxiety and panic disorder comorbidity than patients with correctly identified BD (Matza et al., 2005). This finding demonstrates the complex psychiatric profile of a group of individuals with BD who may be misdiagnosed due to anxiety comorbidities.

Consistent with other studies, (Vojta et al., 2001; Carta et al., 2015; Cooke et al., 1996) we found that patients with BD have similar poor mental and physical functioning on the SF12 as individuals who experience depression. Also consistent with a prior report (Hyun et al., 2000) as well as a higher rate of PTSD diagnosis, we found that physical and sexual abuse as a child or adult were significantly higher in women with BD compared to UD. Psychiatric comorbidity of BD with PTSD is associated with worse physical and mental functioning than individuals with BD alone, which is explained by the severity of associated depressive symptoms (Bajor et al., 2013). The rates of abuse in this sample were 40.6% of women with BD reported physical abuse and 43.8% sexual abuse – twice the rate of women with UD (19.6% and 24.5%, respectively). These rates are similar to the frequency of physical and sexual abuse reported by adult women with BD and MDD of 44% and 25%, respectively (Hyun et al., 2000). Women with BD also were more likely to endorse a greater number of each category of abuse (childhood physical/sexual; adult physical sexual). The experience of abuse in childhood in patients with BD is associated with earlier disease onset, a more severe disease course and poor treatment response (Lippard and Nemeroff, 2020).

Trauma is associated with reduced neuroplasticity (Kauer-Sant’Anna et al., 2007) and increased inflammation (Baumeister et al., 2016). Early childhood traumatic experience is a risk factor for insecure attachment style, which is associated with anxiety, depression and somatization, (Wagner-Skacel et al., 2020) which may be reflected in the significantly higher scores on the SIGH-ADS somatic symptoms-gastrointestinal symptom in women with BD compared to UD. Measures of both maternal sensitivity and mother–infant synchrony were lower in women with BD than in women with UD and without mood disorder (Logsdon et al., 2015). The presence of childhood maltreatment and attachment anxiety may sustain emotion dysregulation and result in greater risk for BD (Kefeli et al., 2018). Some features of depression experienced by postpartum women, including severity, chronicity, anxiety comorbidity, childhood trauma and stress may reflect immunological activation of the innate and adaptive immune responses (Lynall et al., 2020; Felger and Miller, 2020) and with the release of pro-inflammatory biomarkers (Chamberlain et al., 2019). Additional exploration of immunological profiles or immunophenotypes (Felger and Miller, 2020) of depressed postpartum women may be informative in delineating subgroups with inflammation in depression, (Achtyes et al., 2020; Felger and Miller, 2020) or “inflamed depression” (Lynall et al., 2020) and whether this subtype is more frequent in postpartum women with BD compared to UD.

Our investigation is the largest study of symptoms of BD and UD in a perinatal population. Because of the high risk of BD emergence postpartum, improving screening tools is imperative; however, distinguishing BD from UD remains a major challenge to our field (Zimmerman, 2021). All positive screens for BD in postpartum women require additional evaluation with a clinical interview, which is prohibitive in non-mental health settings where most screens occur. A tool that supports confidence that postpartum women who screen negative truly does not have BD (high specificity) would be most useful; however, the sensitivity is likely to be low as is the case with our 8-symptom tool. Recently, investigators studied a small sample of individuals with BD (n = 67) or UD (n = 72) and identified six symptoms and historical characteristics that differentiated the disorders with a sensitivity= 0.88 and specificity = 0.80 (McIntyre et al., 2021). Replication and testing in non-psychiatric settings and in high-risk patients, such as postpartum women, is essential.

4.1. Limitations

This analysis includes women who screened positive on the EPDS, which is a depression screen. Women with hypomania or mania were likely to have had a negative screen at this initial stage and not included in the diagnostic interview. Our study population was an obstetrical population of women who had access to a telephone. A minority of participants completed SCID interviews by telephone instead of an in-person home visit. We included all women who were willing to participate to maximize inclusivity, which outweighed any potential disadvantage that might be inherent in telephone interviews. Postpartum women with unstable living situations, who were hospitalized, or non-responsive to attempts at phone or mail contact did not contribute data.

Acknowledgements

The authors acknowledge Ms. Catalina Montiel and Mrs. Barbara Sutcliffe for managing the citations. We also acknowledge the post-partum women who participated in this investigation.

Funding

Supported by the National Institute of Mental Health: R01 MH071825 (Identification and Therapy of Postpartum Depression, Dr. Wisner); and the National Institute of Child Health and Human Development (K23HD087529, Dr. Clark)

Role of the Funding Source:

The funding source (NIH) played no role in the collection, analysis, interpretation of data, writing of the report or decision to submit the article for publication.

Footnotes

CRediT authorship contribution statement

Crystal T. Clark: Conceptualization, Writing – original draft, Writing – review & editing. Dorothy K. Sit: Investigation, Writing – review & editing. Katelyn B. Zumpf: Data curation, Formal analysis, Writing – review & editing. Jody D. Ciolino: Supervision, Writing – review & editing. Amy Yang: Formal analysis, Writing – review & editing. Sheehan D. Fisher: Writing – review & editing. Katherine L. Wisner: Project administration, Conceptualization, Funding acquisition, Investigation, Methodology, Supervision, Writing – review & editing.

Declaration of Competing Interest

All authors reported no financial interests or potential conflicts of interest.

References

  1. Achtyes E, Keaton SA, Smart L, Burmeister AR, Heilman PL, Krzyzanowski S, Nagalla M, Guillemin GJ, Escobar Galvis ML, Lim CK, Muzik M, Postolache TT, Leach R, Brundin L, 2020. Inflammation and kynurenine pathway dysregulation in post-partum women with severe and suicidal depression. Brain Behav. Immun 83, 239–247. 10.1016/j.bbi.2019.10.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ahrens B, Muller-Oerlinghausen B, Schou M, Wolf T, Alda M, Grof E, Grof P, Lenz G, Simhandl C, Thau K, et al. , 1995. Excess cardiovascular and suicide mortality of affective disorders may be reduced by lithium prophylaxis. J. Affect. Disord 33 (2), 67–75. 10.1016/0165-0327(94)00074-j. [DOI] [PubMed] [Google Scholar]
  3. American College of Obstetricians and Gynecologists. (2018). Screening for perinatal depression ACOG committee opinion,, 132. [DOI] [PubMed] [Google Scholar]
  4. Baethge C, Tondo L, Bratti IM, Bschor T, Bauer M, Viguera AC, Baldessarini RJ, 2003. Prophylaxis latency and outcome in bipolar disorders. Can. J. Psychiatry 48 (7), 449–457. 10.1177/070674370304800704. [DOI] [PubMed] [Google Scholar]
  5. Bajor LA, Lai Z, Goodrich DE, Miller CJ, Penfold RB, Myra Kim H, Bauer MS, Kilbourne AM, 2013. Posttraumatic stress disorder, depression, and health-related quality of life in patients with bipolar disorder: review and new data from a multisite community clinic sample. J. Affect. Disord 145 (2), 232–239. 10.1016/j.jad.2012.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Battle CL, Weinstock LM, Howard M, 2014. Clinical correlates of perinatal bipolar disorder in an interdisciplinary obstetrical hospital setting. J. Affect. Disord 158, 97–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Baumeister D, Akhtar R, Ciufolini S, Pariante CM, Mondelli V, 2016. Childhood trauma and adulthood inflammation: a meta-analysis of peripheral C-reactive protein, interleukin-6 and tumour necrosis factor-alpha. Mol. Psychiatry 21 (5), 642–649. 10.1038/mp.2015.67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Blackmore ER, Rubinow DR, O’Connor TG, Liu X, Tang W, Craddock N, Jones I, 2013. Reproductive outcomes and risk of subsequent illness in women diagnosed with postpartum psychosis. Bipolar Disord. 15 (4), 394–404. 10.1111/bdi.12071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brockington IF, Altman E, Hillier V, Meltzer HY, Nand S, 1982. The clinical picture of bipolar affective disorder in its depressed phase. A report from London and Chicago. Br. J. Psychiatry 141, 558–562. 10.1192/bjp.141.6.558. [DOI] [PubMed] [Google Scholar]
  10. Byatt NDO, Carter DMD, Deligiannidis KMD, Epperson CNMD, Meltzer-Brody SMD, MPH, Payne JMD, Robinson GMD, Silver NMD, Stowe ZMD, Van Niel MSMD, Wisner KMD, Yonkers KMD, 2018. Position Staement on Screening and Treatment of Mood and Anxiety Disorders During Pregnancy and Postpartum. APA Official Actions ed. [Google Scholar]
  11. Carta MG, Norcini-Pala A, Moro MF, Balestrieri M, Caraci F, Dell’Osso L, Sciascio GD, Faravelli C, Hardoy MC, Aguglia E, Roncone R, Nardi AE, Drago F, 2015. Does mood disorder questionnaire identify sub-threshold bipolarity? Evidence studying worsening of quality of life. J. Affect. Disord 183, 173–178. 10.1016/j.jad.2015.04.058. [DOI] [PubMed] [Google Scholar]
  12. Chamberlain SR, Cavanagh J, de Boer P, Mondelli V, Jones DNC, Drevets WC, Cowen PJ, Harrison NA, Pointon L, Pariante CM, Bullmore ET, 2019. Treatment-resistant depression and peripheral C-reactive protein. Br. J. Psychiatry 214 (1), 11–19. 10.1192/bjp.2018.66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Clark CT, Sit DK, Driscoll K, Eng HF, Confer AL, Luther JF, Wisniewski SR, Wisner KL, 2015. Does screening with the MDQ and EPDS improve identification of bipolar disorder in an obstetrical sample? Depress. Anxiety 32 (7), 518–526. 10.1002/da.22373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cooke RG, Robb JC, Young LT, Joffe RT, 1996. Well-being and functioning in patients with bipolar disorder assessed using the MOS 20-ITEM short form (SF-20). J. Affect. Disord 39 (2), 93–97. 10.1016/0165-0327(96)00016-x. [DOI] [PubMed] [Google Scholar]
  15. Cox J, Holden J, 2003. Perinatal Mental Health: a Guide to the Edinburgh Postnatal Depression Scale (EPDS). Royal College of Psychiatrists. [Google Scholar]
  16. Cox JL, Holden JM, Sagovsky R, 1987. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br. J. Psychiatry 150 (6), 782–786. [DOI] [PubMed] [Google Scholar]
  17. Davis N, Smoots A, & Goodman D (2019). Pregnancy-related deaths: data from 14U.S. maternal mortality review committees, 2008–2017. [Google Scholar]
  18. Di Florio A, Forty L, Gordon-Smith K, Heron J, Jones L, Craddock N, Jones I, 2013. Perinatal episodes across the mood disorder spectrum. JAMA Psychiatry 70 (2), 168–175. 10.1001/jamapsychiatry.2013.279. [DOI] [PubMed] [Google Scholar]
  19. Driscoll KE, Sit DKY, Moses-Kolko EL, Pinheiro E, Yang A, Ciolino JD, Eng HF, Luther JF, Clark CT, Wisniewski SR, Wisner KL, 2017. Mood symptoms in pregnant and postpartum women with bipolar disorder: a naturalistic study. Bipolar Disord. 19 (4), 295–304. 10.1111/bdi.12500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. El-Mallakh RS, Vohringer PA, Ostacher MM, Baldassano CF, Holtzman NS, Whitham EA, Thommi SB, Goodwin FK, Ghaemi SN, 2015. Antidepressants worsen rapid-cycling course in bipolar depression: a STEP-BD randomized clinical trial. J .Affect. Disord 184, 318–321. 10.1016/j.jad.2015.04.054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Endicott J, Nee J, Andreasen N, Clayton P, Keller M, Coryell W, 1985. Bipolar II. Combine or keep separate? J. Affect. Disord 8 (1), 17–28. 10.1016/0165-0327(85)90068-0. [DOI] [PubMed] [Google Scholar]
  22. Felger JC, Miller AH, 2020. Identifying Immunophenotypes of Inflammation in Depression: dismantling the Monolith. Biol. Psychiatry 88 (2), 136–138. 10.1016/j.biopsych.2020.04.024. [DOI] [PubMed] [Google Scholar]
  23. Field T, 2010. Postpartum depression effects on early interactions, parenting, and safety practices: a review [doi: 10.1016/j.infbeh.2009.10.005]. Infant Behav. Dev 33, 1–6. https://doi.org/10.1016/j.infbeh.2009.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fritz K, Russell AMT, Allwang C, Kuiper S, Lampe L, Malhi GS, 2017. Is a delay in the diagnosis of bipolar disorder inevitable? Bipolar Disord. 19 (5), 396–400. 10.1111/bdi.12499. [DOI] [PubMed] [Google Scholar]
  25. Frye MA, Calabrese JR, Reed ML, Wagner KD, Lewis L, McNulty J, Hirschfeld RM, 2005. Use of health care services among persons who screen positive for bipolar disorder. Psychiatr. Serv 56 (12), 1529–1533. 10.1176/appi.ps.56.12.1529. [DOI] [PubMed] [Google Scholar]
  26. Gaynes BN, Gavin N, Meltzer-Brody S, Lohr KN, Swinson T, Gartlehner G, Brody S, Miller WC, 2005. Perinatal depression: prevalence, screening accuracy, and screening outcomes: summary. Evid. Rep. Technol. Assess 119, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ghaemi SN, Ko JY, Goodwin FK, 2002. “Cade’s disease” and beyond: misdiagnosis, antidepressant use, and a proposed definition for bipolar spectrum disorder. Can. J. Psychiatry 47 (2), 125–134. 10.1177/070674370204700202. [DOI] [PubMed] [Google Scholar]
  28. Gibson J, McKenzie-McHarg K, Shakespeare J, Price J, Gray R, 2009. A systematic review of studies validating the Edinburgh Postnatal Depression Scale in antepartum and postpartum women. Acta Psychiatr. Scand 119 (5), 350–364. [DOI] [PubMed] [Google Scholar]
  29. Hamilton M, 1960. A rating scale for depression. J. Neurol. Neurosurg. Psychiatr 23 (1), 56–62. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC495331/. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hirschfeld RM, 2002. The mood disorder questionnaire: a simple, patient-rated screening instrument for bipolar disorder. Prim. Care Companion J. Clin. Psychiatry 4 (1), 9–11. 10.4088/pcc.v04n0104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Holm S, 1979. Simple sequentially rejective multiple test procedure. Scand. J. Stat 6 (2), 65–70. [Google Scholar]
  32. Hyun M, Friedman SD, Dunner DL, 2000. Relationship of childhood physical and sexual abuse to adult bipolar disorder. Bipolar Disord. 2 (2), 131–135. 10.1034/j.1399-5618.2000.020206.x. [DOI] [PubMed] [Google Scholar]
  33. Kauer-Sant’Anna M, Tramontina J, Andreazza AC, Cereser K, da Costa S, Santin A, Yatham LN, Kapczinski F, 2007. Traumatic life events in bipolar disorder: impact on BDNF levels and psychopathology. Bipolar Disord. 9 (Suppl 1), 128–135. 10.1111/j.1399-5618.2007.00478.x. [DOI] [PubMed] [Google Scholar]
  34. Kefeli MC, Turow RG, Yildirim A, Boysan M, 2018. Childhood maltreatment is associated with attachment insecurities, dissociation and alexithymia in bipolar disorder. Psychiatry Res. (260), 391–399. [DOI] [PubMed] [Google Scholar]
  35. Kendell RE, Chalmers JC, Platz C, 1987. Epidemiology of puerperal psychoses. Br. J. Psychiatry 150, 662–673. [DOI] [PubMed] [Google Scholar]
  36. Lippard ETC, Nemeroff CB, 2020. The devastating clinical consequences of child abuse and neglect: increased disease vulnerability and poor treatment response in mood disorders. Am. J. Psychiatry 177 (1), 20–36. 10.1176/appi.ajp.2019.19010020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Logsdon MC, Mittelberg M, Jacob AE, Luther JF, Wisniewski SR, Confer A, Eng H, Wisner KL, 2015. Maternal-Infant interaction in women with unipoloar and bipolar depression. Appl. Nurs. Res 28 (4), 381–383. 10.1016/j.apnr.2015.01.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lynall ME, Turner L, Bhatti J, Cavanagh J, de Boer P, Mondelli V, Jones D, Drevets WC, Cowen P, Harrison NA, Pariante CM, Pointon L, Clatworthy MR, Bullmore E, Neuroimmunology of Mood, D., Alzheimer’s Disease, C., 2020. Peripheral blood cell-stratified subgroups of inflamed depression. Biol. Psychiatry 88 (2), 185–196. 10.1016/j.biopsych.2019.11.017. [DOI] [PubMed] [Google Scholar]
  39. Mahmoud DR, Yang A, Ciolino JD, Fisher SD, Sit D, Pinheiro E, Pendergrast T, O’Shea K, Wisner KL, Clark CT, 2019. Validity of the WHIPLASHED as a tool to identify bipolar disorder in women. J. Affect. Disord 246, 69–73. 10.1016/j.jad.2018.12.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Masters GA, Brenckle L, Sankaran P, Person SD, Allison J, Moore Simas TA, Ko JY, Robbins CL, Marsh W, Byatt N, 2019. Positive screening rates for bipolar disorder in pregnant and postpartum women and associated risk factors. Gen. Hosp. Psychiatry 61, 53–59. 10.1016/j.genhosppsych.2019.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Matza LS, Rajagopalan KS, Thompson CL, de Lissovoy G, 2005. Misdiagnosed patients with bipolar disorder: comorbidities, treatment patterns, and direct treatment costs. J. Clin. Psychiatry 66 (11), 1432–1440. 10.4088/jcp.v66n1114. [DOI] [PubMed] [Google Scholar]
  42. McIntyre RS, Patel MD, Masand PS, Harrington A, Gillard P, McElroy SL, Sullivan K, Montano CB, Brown TM, Nelson L, Jain R, 2021. The Rapid Mood Screener (RMS): a novel and pragmatic screener for bipolar I disorder. Curr. Med. Res. Opin 37 (1), 135–144. 10.1080/03007995.2020.1860358. [DOI] [PubMed] [Google Scholar]
  43. Merikangas KR, Akiskal HS, Angst J, Greenberg PE, Hirschfeld RM, Petukhova M, Kessler RC, 2007. Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication [Comparative Study Research Support, N I H, Extramural Research Support, Non-U S Gov’t]. Arch. Gen. Psychiatry 64 (5), 543–552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Merrill L, Mittal L, Nicoloro J, Caiozzo C, Maciejewski PK, Miller LJ, 2015. Screening for bipolar disorder during pregnancy. Arch. Womens Ment. Health 18 (4), 579–583. 10.1007/s00737-015-0527-y. [DOI] [PubMed] [Google Scholar]
  45. Mitchell P, Parker G, Jamieson K, Wilhelm K, Hickie I, Brodaty H, Boyce P, Hadzi-Pavlovic D, Roy K, 1992. Are there any differences between bipolar and unipolar melancholia? J. Affect. Disord 25 (2), 97–105. 10.1016/0165-0327(92)90072-e. [DOI] [PubMed] [Google Scholar]
  46. Mitchell PB, Wilhelm K, Parker G, Austin MP, Rutgers P, Malhi GS, 2001. The clinical features of bipolar depression: a comparison with matched major depressive disorder patients. J. Clin. Psychiatry 62 (3), 212–216 quiz 217. https://www.ncbi.nlm.nih.gov/pubmed/11305713. [PubMed] [Google Scholar]
  47. Moses-Kolko EL, Price JC, Wisner KL, Hanusa BH, Meltzer CC, Berga SL, Grace AA, di Scalea TL, Kaye WH, Becker C, Drevets WC, 2012. Postpartum and depression status are associated with lower [[(1)(1)C]raclopride BP(ND) in reproductive-age women. Neuropsychopharmacology 37 (6), 1422–1432. 10.1038/npp.2011.328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Munk-Olsen T, Laursen TM, Pederson CB, Mors O, Mortensen PB, 2006. New parents and mental disorders: a population-based register study. JAMA 296 (21), 2582–2589. 10.1001/jama.296.21.2582. [DOI] [PubMed] [Google Scholar]
  49. Munk-Olsen T, Laursen TM, Meltzer-Brody S, Mortensen PB, Jones I, 2012. Psychiatric disorders with postpartum onset: possible early manifestations of bipolar affective disorders. Arch. Gen. Psychiatry 69 (4), 428–434. 10.1001/archgenpsychiatry.2011.157. [DOI] [PubMed] [Google Scholar]
  50. Parker G, Roy K, Wilhelm K, Mitchell P, Hadzi-Pavlovic D, 2000. The nature of bipolar depression: implications for the definition of melancholia. J. Affect. Disord 59 (3), 217–224. 10.1016/s0165-0327(99)00144-5. [DOI] [PubMed] [Google Scholar]
  51. Perlis RH, Brown E, Baker RW, Nierenberg AA, 2006. Clinical features of bipolar depression versus major depressive disorder in large multicenter trials. Am. J. Psychiatry 163 (2), 225–231. 10.1176/appi.ajp.163.2.225. [DOI] [PubMed] [Google Scholar]
  52. Perugi G, Akiskal HS, Lattanzi L, Cecconi D, Mastrocinque C, Patronelli A, Vignoli S, Bemi E, 1998. The high prevalence of “soft” bipolar (II) features in atypical depression. Compr. Psychiatry 39 (2), 63–71. 10.1016/s0010-440x(98)90080-3. [DOI] [PubMed] [Google Scholar]
  53. Sharma V, Khan M, Corpse C, Sharma P, 2008. Missed bipolarity and psychiatric comorbidity in women with postpartum depression. Bipolar Disord. 10 (6), 742–747. 10.1111/j.1399-5618.2008.00606.x. [DOI] [PubMed] [Google Scholar]
  54. Siu AL, Force, U.S.P.S.T., Bibbins-Domingo K, Grossman DC, Baumann LC, Davidson KW, Ebell M, Garcia FA, Gillman M, Herzstein J, Kemper AR, Krist AH, Kurth AE, Owens DK, Phillips WR, Phipps MG, Pignone MP, 2016. Screening for depression in adults: us preventive services task force recommendation statement. JAMA 315 (4), 380–387. 10.1001/jama.2015.18392. [DOI] [PubMed] [Google Scholar]
  55. Vojta C, Kinosian B, Glick H, Altshuler L, Bauer MS, 2001. Self-reported quality of life across mood states in bipolar disorder. Compr. Psychiatry 42 (3), 190–195. 10.1053/comp.2001.23143. [DOI] [PubMed] [Google Scholar]
  56. Wagner-Skacel J, Bengesser S, Dalkner N, Morkl S, Painold A, Hamm C, Pilz R, Rieger A, Kapfhammer HP, Hiebler-Ragger M, Jauk E, Butler MI, Reininghaus EZ, 2020. Personality structure and attachment in bipolar disorder. Front. Psychiatry 11, 410. 10.3389/fpsyt.2020.00410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Ware J Jr., Kosinski M, Keller SD, 1996. A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med. Care 34 (3), 220–233. [DOI] [PubMed] [Google Scholar]
  58. Williams J, Terman M, 2003. Structured Interview Guide for the Hamilton Depression Rating Scale With Atypical Depression Supplement (SIGH-ADS). New York State Psychiatric Institute, New York. [Google Scholar]
  59. Wisner KL, Sit DK, McShea MC, Rizzo DM, Zoretich RA, Hughes CL, Eng HF, Luther JF, Wisniewski SR, Costantino ML, 2013. Onset timing, thoughts of selfharm, and diagnoses in postpartum women with screen-positive depression findings. JAMA Psychiatry 70 (5), 490–498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zimmerman M, 2021. Using screening scales for bipolar disorder in epidemiologic studies: lessons not yet learned. J. Affect. Disord 292, 708–713. 10.1016/j.jad.2021.06.009. [DOI] [PubMed] [Google Scholar]

RESOURCES