The American Psychiatric Association (APA) has updated its Privacy Policy and Terms of Use, including with new information specifically addressed to individuals in the European Economic Area. As described in the Privacy Policy and Terms of Use, this website utilizes cookies, including for the purpose of offering an optimal online experience and services tailored to your preferences.

Please read the entire Privacy Policy and Terms of Use. By closing this message, browsing this website, continuing the navigation, or otherwise continuing to use the APA's websites, you confirm that you understand and accept the terms of the Privacy Policy and Terms of Use, including the utilization of cookies.

×
ArticlesFull Access

Change Mechanisms in Brief Interpersonal Psychotherapy for Women With Perinatal Depression: Qualitative Study

Abstract

Objective:

Brief interpersonal psychotherapy (IPT-B) has been shown to be effective in treating perinatal depression and in preventing depressive relapse among socioeconomically disadvantaged women. Yet, it is unclear how IPT-B alleviates depression. Previous research has suggested four possible change mechanisms derived from IPT’s interpersonal model: decreasing interpersonal stress, facilitating emotional processing, improving interpersonal skills, and enhancing social support. This study explored how women who received IPT-B or enhanced maternity support services (MSS-Plus) evaluated their respective experiences.

Methods:

A qualitative study was conducted with 16 women who had been recruited from public health clinics to participate in a larger, randomized controlled trial of women with major depression or dysthymia and who had been assigned to receive IPT-B or MSS-Plus. The sample was 63% non-Hispanic White, had an average age of 31.6 years, and was balanced in intervention group assignment, posttraumatic stress disorder status, and depression improvement. Telephone interviews included semistructured, open-ended questions eliciting participants’ experiences with depression treatment. Predetermined, conceptually derived codes were based on the four postulated IPT change mechanisms.

Results:

Thematic coded excerpts were collected and discussed. Excerpts lent support to the role of IPT-B in helping women decrease their interpersonal stress; identify, reflect on, and regulate their emotions; and improve their social skills. Evidence for increasing social support was mixed but highlighted the importance of the therapeutic relationship.

Conclusions:

Including qualitative findings into training in public health and other clinical settings will help illuminate the role of the provider in facilitating the change mechanisms that may lead to improved mental health among clients.

Highlights

  • The qualitative findings of this study support the role of brief interpersonal psychotherapy (IPT-B) in activating possible mechanisms of change among women with perinatal depression.

  • These qualitative results showed that IPT-B, compared with enhanced maternity support services (MSS-Plus), helped women with perinatal depression to decrease interpersonal stress; improve interpersonal skills; and identify, reflect on, and regulate their emotions.

  • Women who received IPT-B reported feeling more connected to their therapist as a transitional source of social support compared with those who received MSS-Plus.

Perinatal depression is a serious mental health condition, with a prevalence ranging from 25% to 30% among socioeconomically disadvantaged women in the United States—about twice the rate of middle- and upper-income women (1, 2). Posttraumatic stress disorder (PTSD), a chronic debilitating psychiatric disorder, occurs among women at a rate of 10.4%, nearly double that found among men (3), and is often comorbid with major depressive disorder, particularly for those of lower socioeconomic status (47). A randomized controlled trial (RCT) demonstrated the effectiveness of collaborative care for perinatal depression, including brief interpersonal psychotherapy (IPT-B), compared with enhanced public health maternity support services (MSS-Plus) in the Seattle–King County public health system, from pregnancy up to 15 months postpartum, for socioeconomically disadvantaged women with major depression and/or dysthymia, 65% of whom had probable comorbid PTSD (8). The RCT results indicated that, compared with MSS-Plus, IPT-B significantly reduced depressive symptoms, achieved higher remission rates, and resulted in higher quality of care across 18 months postbaseline (9). Furthermore, women with comorbid PTSD receiving IPT-B compared with MSS-Plus showed significantly greater reductions in depression severity and PTSD severity, more improved functioning, higher rates of depression response and remission, and increased use of mental health services and antidepressant medication (10).

What we do not know from these quantitative findings, however, is how the women who received IPT-B or MSS-Plus viewed and valued their respective experiences after the trial was over. Ascertaining the participants’ qualitative perspectives on IPT-B and MSS-Plus is important for identifying the change mechanisms that facilitated amelioration of perinatal depression and improved functioning. Thus, this qualitative study aimed to explore which aspects of IPT-B and MSS-Plus our participants found helpful and to what extent they felt empowered by the end of the study. In fact, patients with depression may consider symptom remission as only one factor in determining their state of remission and may view the positive aspects of mental health, such as optimism, feeling in control of emotions, and participating in and enjoying relationships with family and friends, as equally, if not more important, than the absence of depressive symptoms (11).

In this study, we used the framework method of qualitative analysis with a deductive approach (12), primarily to explore how participants receiving IPT-B or MSS-Plus viewed their respective treatment experiences in light of IPT’s conceptual model of change postulated by Lipsitz and Markowitz (13), which we describe below. The framework method is not aligned with a particular epistemological viewpoint and has no allegiance to an inductive or deductive approach, but it provides a structure into which researchers can systematically reduce qualitative data in order to analyze them by case and code. As Gale et al. (12) have suggested, a deductive approach is appropriate when there are specific conceptual questions to explore and comparisons to be made across groups of cases. In the deductive approach, themes and codes are preselected based on a previous theory. Thus, the qualitative researcher retrieves data coded with both a conceptual code and a participant code (for example, IPT-B or MSS-Plus) and compares whether certain conceptual themes are more apparent in one group than another (14).

The conceptual model guiding the derivation of our thematic codes was based on IPT’s interpersonal model, which integrates two interpersonal frameworks (13). The first framework encompasses relational theory (15) and attachment theory (16), both of which insist that mental health and attachment depend on healthy, intimate connections with other people and remain essential throughout life in providing individuals with nurturance and a secure base, especially under conditions of stress. The second framework includes the considerable research on stress, social support, and illness (1719), which informs IPT’s focus on identifying the important life events influencing the onset and course of illness (20) and on resolving the current interpersonal context reciprocally linked with the illness (13).

In postulating how IPT helps alleviate depression through resolving the key interpersonal problems most linked with distress, Lipsitz and Markowitz (13) proposed four mechanisms of change derived from IPT’s interpersonal model: decreasing interpersonal stress, facilitating emotional processing, improving interpersonal skills, and enhancing social support. We therefore developed codes to reflect these mechanisms of change in assessing participants’ responses to open-ended questions about their experiences with IPT-B or MSS-Plus. To date, limited qualitative research has explored the change mechanisms of IPT, despite the therapy’s established efficacy and effectiveness. One exception is a study that conducted a thematic analysis of 10 sets of audio-recorded IPT sessions to explore the process of change and found that the characteristics associated with response to IPT involved an ability to engage in multiple perspectives, an awareness of others’ feelings, an ability to use a range of communication strategies, and an ability to engage with the therapist (21). However, these findings were based on a single case with no comparison sample.

To adequately capture participants’ views of their treatment experiences (22), in our qualitative study we sought to provide a quantitative summary of women’s thematic endorsements of these change mechanisms. As Sandelowski (23) has argued, numbers may be integral to qualitative research, because meaning depends, in part, on number. The latter point has been illustrated by qualitative research in which responses about depression treatment have been counted (24). On the basis of Lipsitz and Markowitz’s work (13), we expected that participants assigned to receive IPT-B would be more likely than those assigned to receive MSS-Plus to mention the four hypothesized change mechanisms in their responses to our qualitative questions.

Methods

The MOMCare IPT-B RCT recruited participants from 10 clinics in the public health system of Seattle–King County and served 164 Medicaid-covered pregnant women with major depression and/or dysthymia; 65% of the participants had probable comorbid PTSD. The women were randomly assigned to receive either IPT-B or MSS-Plus. The latter was provided by public health social workers who delivered case management plus referral to community mental health services for women from pregnancy through 2 months postpartum. The study intervention was conducted from January 2010 through November 2013. Participants completed structured research interviews at baseline and at 3, 6, 12, and 18 months after baseline. Details of the RCT have been published (810).

After the RCT was completed, we conducted a qualitative study (February 2014 through August 2014) with 16 participants. The qualitative study and the RCT were reviewed and approved by the institutional review board at the University of Washington. All participants provided verbal informed consent for their participation and for audio recording of the qualitative study’s phone interviews. Average time from the last RCT interview to the qualitative study interview was 1.4 years.

Participants

We used stratified purposive sampling based on three factors expected to influence participants’ views of their treatment experiences. The goal of stratified purposive sampling is to sample a limited number of participants for in-depth analysis that can best help the qualitative researcher understand the central question under study (25). The sample for the qualitative study was selected for balance among three factors: intervention (IPT-B or MSS-Plus), comorbid PTSD status (with PTSD or without PTSD), and depression improvement at 12 months after baseline (low or high improvement). Depression improvement was defined as the percentage change in depression difference, measured by Symptom Checklist–20 (26) depression scores, from baseline to 12 months postbaseline, with scores in the top tertile representing high improvement and scores in the bottom tertile representing low improvement. All eligible participants were categorized into one of eight subgroups: IPT-B, low depression improvement, and PTSD; IPT-B, high depression improvement, and PTSD; IPT-B, low depression improvement, no PTSD; IPT-B, high depression improvement, no PTSD; or the same subgroup patterns for those assigned to MSS-Plus. Participants were randomly ordered within each subgroup.

The qualitative interviewer made phone calls starting with the first randomly ordered participant for each of the eight subgroups and continued until she reached a participant. The goal was to stop after the interviewer had spoken with two participants in each of the eight subgroups if responses became repetitive and saturation was reached. Because of delays in hearing back from some participants and our desire to include all women who were interested in participating, the final sample had a slight imbalance on depression improvement; the MSS-Plus group had one more participant with high depression improvement than the IPT-B group, and the IPT-B group had one more participant with low depression improvement than the MSS-Plus group. Despite the slight imbalance, responses had become repetitive and we determined that saturation had been reached, so we stopped enrollment with 16 completed interviews.

Qualitative Interviews

Through an iterative process that included practice interviews and team discussion of responses, the research team first developed a standard set of semistructured, open-ended questions and probes crafted to explore what participants in IPT-B and MSS-Plus viewed as change mechanisms in their treatment experiences with the IPT-B therapist or other professional helper (see the online supplement for the interview questions). A standard set of questions was asked of each participant; however, probes to clarify responses could be individually tailored. To preserve the integrity of the RCT and to avoid making unplanned modifications resulting from the qualitative study, we conducted the interviews after the RCT was completed (27).

For the qualitative study, a single interviewer (with a master’s degree in social work and decades of experience conducting qualitative interviews) used the questions developed to interview the 16 participants. The interviewer was not blind to group assignment because she needed to track participation by category and some of the follow-up probes varied by intervention group. Phone interviews averaged 34 minutes, and participants received a $30 gift card for their time.

Coding

The phone interviews were audio recorded, transcribed, deidentified, and uploaded to the Dedoose (28) web application for managing, analyzing, and presenting qualitative and mixed methods research data. The research team developed thematic codes based on the first two IPT-B interviews and the first two MSS-Plus interviews. We used the framework method (12), with predetermined, conceptually derived codes that were based on the four identified change mechanisms (13). In reviewing the transcripts, we decided to categorize transcripts as indicative of “decreasing interpersonal stress” when the participant talked about feeling less stressed or successfully managing an interpersonal problem to reduce stress. “Facilitating emotional processing” was noted when a participant expressed the ability to identify her emotions; to reflect on her emotions, especially complex ones, as the causes of or contributors to the problem; or to regulate her emotions. “Improvement of interpersonal skills” was noted when a participant talked about being able to calmly express her emotions to others or to speak up assertively for what she needed or wanted. “Enhancing social support” was noted when a participant reported engaging with and receiving pleasure or support from her family, husband or partner, friends, baby and/or children, work colleagues, or a therapist or service provider (because the IPT model views the therapeutic relationship as a transitional source of social support) (13).

Two primary coders created excerpts from the transcripts and assigned codes for each of the 16 interviews. Excerpts could be coded as supporting multiple themes. All transcripts were independently coded by both primary coders. The coders were not blind to intervention condition; however, they were unaware of comorbid PTSD and depression improvement status. A third research team member compared and verified the codes, finding excerpts that the two primary coders had not agreed on in identifying or coding. A fourth team member joined the group meetings to discuss and review all discrepant or inconsistent codes. If consensus could not be reached among the four research team members or if the team determined that the excerpt did not reflect a theme, the excerpt was omitted from the analysis; 17% of the excerpts (N=68) were omitted. In line with the framework method’s use of a deductive approach (12), we based our data interpretation on a quantitative summary (i.e., counts) of the women’s thematic coded excerpts from IPT-B and MSS-Plus. Participant responses are provided below for illustrative purposes.

Results

At the time of the qualitative study, the mothers’ average age was 31.6 years; 63% of our sample were non-Hispanic White. Table 1 shows the distribution of depression improvement and comorbid PTSD among the women in the qualitative sample, by type of intervention. Table 2 presents the demographic and clinical baseline characteristics of the qualitative sample and the full RCT sample, including major depression (29), dysthymia (30), and PTSD (31). The qualitative sample did not significantly differ from the RCT sample on the variables examined. Table 3 presents the number of excerpts reflecting the change mechanism themes for the IPT-B and MSS-Plus groups.

TABLE 1. Grouping of qualitative sample (N=16) by treatment group, depression improvement, and PTSD statusa

SubgroupPTSDNo PTSD
IPT-B group
 Low improvement31
 High improvement22
MSS-Plus group
 Low improvement21
 High improvement32

aIPT-B, brief interpersonal psychotherapy; MSS-Plus, enhanced public health maternity support services; PTSD, posttraumatic stress disorder.

TABLE 1. Grouping of qualitative sample (N=16) by treatment group, depression improvement, and PTSD statusa

Enlarge table

TABLE 2. Demographic and clinical characteristics of the qualitative sample and full RCT sample at baselinea

Qualitative sample (N=16)RCT full sample (N=164)
CharacteristicN%N%
Intervention group
 IPT-B8508149
 MSS-Plus8508351
Race-ethnicity
 White10636741
 African American3193924
 Latina2133823
 Asian/Pacific Islander16127
 American Indian/Alaska Native0085
Partner status
 Married3194829
 Living with a partner6385534
 Partner, not living with2132113
 No partner5314024
Education
 Less than high school4253521
 High school degree/GED3193421
 Some college/vocational8507747
 College degree or more161811
Employed5315735
Annual income ≤$10,0006406742
Unplanned pregnancy148811872
Depressive disorders
 Major depression disorderb1610015896
 Dysthymiac3193924
 Major depression and dysthymia3193421
Posttraumatic stress disorderd106310665

aIPT-B, brief interpersonal psychotherapy; MSS-Plus, enhanced maternity support services; RCT, randomized controlled trial.

bMeasured with the Patient Health Questionnaire depression scale (29).

cMeasured with the MINI-International Neuropsychiatric Interview dysthymia module (30).

dMeasured with the Post-Traumatic Stress Disorder Checklist-Civilian Version (31).

TABLE 2. Demographic and clinical characteristics of the qualitative sample and full RCT sample at baselinea

Enlarge table

TABLE 3. Number of interview excerpts endorsing change mechanisms, by intervention typea

ThemeIPT-BMSS-Plus
Decreasing interpersonal stress
Dealing with an interpersonal problem or reducing interpersonal stress (actively dealing with it)186
Facilitating emotional processing
Becoming more able to identify one’s own emotions319
Becoming more able to reflect on one’s own emotional experiences as to what might have caused or contributed to their depression (link events to symptoms)4124
Becoming more able to regulate one’s own emotions114
Improving interpersonal skills
Communicating more effectively or assertively with others155
Enhancing social support
Engaging with or receiving pleasure or support from one’s family, friends, work colleagues, baby, other children2330
Engaging with and feeling connected to a therapist or service provider349

aIPT-B, brief interpersonal psychotherapy; MSS-Plus, enhanced maternity support services. The codebook is available from the authors on request.

TABLE 3. Number of interview excerpts endorsing change mechanisms, by intervention typea

Enlarge table

Decreasing Interpersonal Stress

Participants in the IPT-B group had more coded excerpts (N=18) about successfully dealing with an interpersonal problem or reducing interpersonal stress than did those in the MSS-Plus group (N=6). The women often mentioned conflicts and stress in their relationships with their partners, people at work, and family members. Success in managing these stressors sometimes came from accepting the limitations of the relationship (“I decided to expect less and live with him the way he is”) or by reframing and seeing the relationship in another way:

I was gonna leave my boyfriend and go to someone else, go to another man. But [the therapist] helped me see that I didn’t always talk bad about him, there were good things about him. So, you know, she helped me see the good was better than the bad. . . . And I’m glad I didn’t, cause it would’ve split my family up.

In some cases, the stress was managed by discontinuing the relationship: “Well, a lot of the reason is most of my anger was towards my baby daddy. . . and we’re separated now, so I think that’s honestly one of the main things that I needed to do.”

Some women mentioned learning skills or feeling supported to actively do something to solve the problem:

I think during the time when my family doesn’t know that I am pregnant, and I’m so scared to tell them. And then [the therapist] is there, and she asked me when am I ready to tell them, and then she was so helpful making me comfortable telling them.

Another woman commented,

At work, it’s totally different. They were totally shocked. Before they told me, “You’re so quiet, you don’t complain, you just do your job, whatever they tell you to do, you do it,” and now . . . I can tell them whatever, not really whatever, it’s just that I can tell them how I feel.

Facilitating Emotional Processing

Participants in the IPT-B group, compared with those in MSS-Plus group, had more coded excerpts that reflected becoming more able to identify their emotions (31 versus 9, respectively), more able to reflect on their emotional experiences (41 versus 24, respectively), and more able to regulate their emotions (11 versus 4, respectively). Excerpts mentioning “feeling” words, such as “I felt angry and frustrated” and “I felt so sad and defeated when he left me,” were coded as identifying emotions.

Excerpts were coded as indicative of increased ability to reflect on emotional experiences if they linked the event or situation that might have caused or contributed to the emotions, feelings, or depression. One participant said, “So [my baby’s father] was living with us and drinking a lot, working less and less. It was just all this big snowball effect, and it was probably one of the lowest points in my life.” Another participant commented, “Every time from work, I came home, and I’m so stressed, and then I get mad a lot. And then my kids are telling me ‘Mom, why are you mad? We didn’t do anything.’ It’s because I couldn’t express my feelings at work.”

Examples of participants becoming more able to regulate their emotions included, “I look at what’s going on and really determine if it’s worth getting so upset about and try and slow down and do it one step at a time.” Another participant expressed, “Every time I blew up it’s just, you know, everything comes out. I’m mad and everything. Now I can control it . . . especially how I feel.”

Improving Interpersonal Skills

Participants in the IPT-B group had more excerpts indicating improved interpersonal skills than did those in the MSS-Plus group (15 versus 5, respectively). Examples of excerpts indicating communicating more effectively or assertively with others included the following: “Before, I was so scared to tell people how I think. . . that I don’t want to hurt their feelings and everything. Cause I’ve been doing that for so long, and then now I can. I know how to express myself without hurting somebody.” Another participant said, “I’m more confident now, on telling people how I feel—not like before [when] I [kept] it to myself.”

Enhancing Social Support

Results were mixed for enhanced social support. Participants in the IPT-B group did not show more endorsements than those in the MSS-Plus group of engaging with or receiving pleasure or support from their family, friends, work colleagues, baby, or other children (23 versus 30, respectively). Statements indicative of enhanced social support included, “Well, my friends kept me busy, that’s for sure” and “My family made sure that I was not alone. They kept coming over.”

The IPT-B group, however, had more endorsements (N=34) of receiving positive support from and feeling connected to their therapist or service provider, compared with the MSS-Plus group (N=9). Many women were socially isolated, and the therapist became a transitional social support person for them:

I just needed someone to listen, and she was there. I know it’s hard for me when somebody’s going through something. It’s kind [of] hard for me to say anything to them except, “I’m sorry.” And she was just really good with not saying “I’m sorry,” but she would talk to me about it. And I guess I’ve never had somebody like that, not even a friend or a family member.

Another participant said,

I was really depressed at the time, because I didn’t have my family around me when I was pregnant. I was really scared to talk to my boyfriend about stuff, and then I was scared to talk to my mom about stuff. So, I could talk to [the therapist] about anything and everything, and she wouldn’t judge me or anything. She’s just someone who will listen to me, cause I don’t have friends or anything. I kind of considered her a friend, cause I confided in her.

Women in the IPT-B group made many comments about appreciating the support of the therapist and had positive comments about the intervention. “I feel like a lot of people give up on me, too. And my therapist did not want to give up on me,” said one participant. Another commented, “Then I was just really depressed, and kind [of] like down in the dumps all the time. And now I’m happy, I’m working, I have stuff to do. I take care of my 2-year-old.”

Discussion

These qualitative findings provide support for the role of IPT-B in activating the conceptually derived change mechanisms postulated by Lipsitz and Markowitz (13). Specifically, we found that IPT-B, compared with MSS-Plus, was more effective in helping socioeconomically disadvantaged women to decrease interpersonal stress, improve interpersonal skills, and process their emotions. The latter involved becoming more able to identify their emotions, reflect on their emotions, and regulate their emotions. Finally, women in the IPT-B group reported more connection to their therapist as a transitional source of social support.

We did not find evidence that IPT-B enhanced social support from family and friends at a higher rate than MSS-Plus. Although the women in the IPT-B group endorsed engaging with their support networks, it was not with the same frequency as was mentioned by those in the MSS-Plus group. These results for social support may not be surprising, because the MSS-Plus providers deliberately aimed to connect pregnant women with a social support network. The women in the MSS-Plus group generally did not report having a therapeutic relationship with their provider and primarily received case management and referrals; thus, they may have spent most of the time in the qualitative interview discussing those experiences, resulting in more coded excerpts related to enhancing social support.

These qualitative viewpoints on the importance of facilitating emotional processing and improving interpersonal skills in IPT-B during the perinatal period correspond well with research on emotional intelligence (32). Emotional intelligence is defined as the ability to accurately and adaptively perceive, use, understand, and manage emotions—skills that can be enhanced and taught to produce socially relevant outcomes (33). The theoretical model of emotional intelligence, like that of IPT, emphasizes that emotions convey important information as communication signals that play a key role in helping people understand their own and others’ emotions and actions. It is not surprising that higher scores on emotional intelligence correlate with reduced dysthymia (34) and increased social competence and interpersonal sensitivity (35). In sum, it is possible that IPT-B in the MOMCare trial enhanced some key aspects of emotional intelligence among the participants.

Although the results of the study are promising, this research had some limitations. Because the qualitative interviews occurred after the RCT intervention was complete, the participants had to recall events from, on average, over 1 year prior. The qualitative interviewer was skilled in active listening and allowed the participants time to recall their experiences; however, memories can fade over time. Also, the participants may have tended to remember the more positive experiences of the intervention and may have wanted to please the interviewer, even though they were assured that we wanted to learn about good as well as bad experiences. This tendency could have affected the data regarding their feelings toward their therapist. However, the fact that they used more language reflecting the change mechanisms attributed to IPT-B more than a year, on average, after their intervention was completed speaks to the potential lasting impact of this therapy. Finally, the study was limited because neither the interviewer nor the coders for the qualitative study were blind to treatment condition, which could have led to biased interview responses by the participants or to biased coding of participant responses.

Conclusions

In conclusion, this qualitative study has contributed to the literature by exploring how women who received IPT-B or MSS-Plus for perinatal depression evaluated their respective experiences after the intervention was completed. The aim of our analysis was to augment the quantitative results from our RCT with salient qualitative themes to inform the integration of stepped-care treatment for high-risk pregnant and postpartum women with major depression, with and without comorbid PTSD, in U.S. public health systems. These qualitative findings thus complement the established quantitative results on the value of IPT for socioeconomically disadvantaged women during the perinatal period (9, 10).

Future quantitative and qualitative research should involve testing the four identified change mechanisms as mediators of the effect of IPT-B on depression amelioration. To further disseminate IPT-B into public health or other clinical settings, it will be important to include client voices into IPT-B training so that providers will understand their own important role in facilitating the change mechanisms that lead to improved mental health for their clients.

School of Social Work (Grote, Lohr, Curran), Department of Child, Family, and Population Health Nursing (Lohr), and Department of Rehabilitation Medicine (Curran), University of Washington, Seattle; Department of Social Work, Seattle University, Seattle (Cristofalo).
Send correspondence to Ms. Lohr ().

This study was presented in part at the annual meeting of the Society for Social Work and Research, January 11, 2018, Washington, D.C., and at the Eighth Biennial Congress of the International Society of Interpersonal Psychotherapy, November 9, 2019, Budapest.

This study was funded by the National Institute of Mental Health (R01-MH084897; principal investigator, Dr. Grote).

The authors report no financial relationships with commercial interests.

The authors thank Marilyn Gregory for conducting the interviews, Elaine Cartwright Howell for transcription services, Molly Klekamp for coding, and Wayne J. Katon for support and inspiration.

References

1 Hobfoll SE, Ritter C, Lavin J, et al.: Depression prevalence and incidence among inner-city pregnant and postpartum women. J Consult Clin Psychol 1995; 63:445–453Crossref, MedlineGoogle Scholar

2 Canady RB, Bullen BL, Holzman C, et al.: Discrimination and symptoms of depression in pregnancy among African American and White women. Womens Health Issues 2008; 18:292–300Crossref, MedlineGoogle Scholar

3 Breslau N, Davis GC, Peterson EL, et al.: Psychiatric sequelae of posttraumatic stress disorder in women. Arch Gen Psychiatry 1997; 54:81–87Crossref, MedlineGoogle Scholar

4 Kessler RC, Sonnega A, Bromet E, et al.: Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry 1995; 52:1048–1060Crossref, MedlineGoogle Scholar

5 Brown TA, Campbell LA, Lehman CL, et al.: Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. J Abnorm Psychol 2001; 110:585–599Crossref, MedlineGoogle Scholar

6 Seng JS, Rauch SA, Resnick H, et al.: Exploring posttraumatic stress disorder symptom profile among pregnant women. J Psychosom Obstet Gynaecol 2010; 31:176–187Crossref, MedlineGoogle Scholar

7 Seng JS, Low LK, Sperlich M, et al.: Prevalence, trauma history, and risk for posttraumatic stress disorder among nulliparous women in maternity care. Obstet Gynecol 2009; 114:839–847Crossref, MedlineGoogle Scholar

8 Grote NK, Katon WJ, Lohr MJ, et al.: Culturally relevant treatment services for perinatal depression in socio-economically disadvantaged women: the design of the MOMCare study. Contemp Clin Trials 2014; 39:34–49Crossref, MedlineGoogle Scholar

9 Grote NK, Katon WJ, Russo JE, et al.: Collaborative care for perinatal depression in socioeconomically disadvantaged women: a randomized trial. Depress Anxiety 2015; 32:821–834Crossref, MedlineGoogle Scholar

10 Grote NK, Katon WJ, Russo J, et al.: A randomized trial of collaborative care for perinatal depression in socioeconomically disadvantaged women: the impact of comorbid posttraumatic stress disorder. J Clin Psychol 2016; 77:1527–1537Google Scholar

11 Zimmerman C: Trafficking in Women: A Qualitative Study to Conceptualise and Map Health Risks. Doctoral thesis. London School of Hygiene & Tropical Medicine, Public Health Policy Unit, London, 2006Google Scholar

12 Gale NK, Heath G, Cameron E, et al.: Using the framework method for the analysis of qualitative data in multidisciplinary health research. BMC Med Res Methodol 2013; 13:117Crossref, MedlineGoogle Scholar

13 Lipsitz JD, Markowitz JC: Mechanisms of change in interpersonal psychotherapy (IPT). Clin Psychol Rev 2013; 33:1134–1147Crossref, MedlineGoogle Scholar

14 Bradley EH, Curry LA, Devers KJ: Qualitative data analysis for health services research: developing taxonomy, themes, and theory. Health Serv Res 2007; 42:1758–1772Crossref, MedlineGoogle Scholar

15 Sullivan HS: The Interpersonal Theory of Psychiatry. New York, WW Norton, 1953Google Scholar

16 Bowlby J: Attachment and Loss. London, Hogarth, 1969Google Scholar

17 Brown GW, Harris T: Social Origins of Depression: A Study of Psychiatric Disorder in Women. Oxon, UK, Taylor & Francis, 1978Google Scholar

18 Cohen S, Wills TA: Stress, social support, and the buffering hypothesis. Psychol Bull 1985; 98:310–357Crossref, MedlineGoogle Scholar

19 Paykel ES, Myers JK, Dienelt MN, et al.: Life events and depression. A controlled study. Arch Gen Psychiatry 1969; 21:753–760Crossref, MedlineGoogle Scholar

20 Winters EE (ed): The Collected Papers of Adolf Meyer, vol. 2. Baltimore, MD, Johns Hopkins, 1951Google Scholar

21 Crowe M, Luty S: Patterns of response and nonresponse in interpersonal psychotherapy: a qualitative study. Psychiatry 2005; 68:337–349Crossref, MedlineGoogle Scholar

22 Lofland J, Snow D, Anderson L, et al.: Analyzing Social Settings: A Guide to Qualitative Observation and Analysis, 4th ed. Belmont, CA, Wadsworth Thompson, 2006Google Scholar

23 Sandelowski M: Real qualitative researchers do not count: the use of numbers in qualitative research. Res Nurs Health 2001; 24:230–240Crossref, MedlineGoogle Scholar

24 Orengo-Aguayo RE, Segre LS: Depression treatment delivered at the point-of-care: a qualitative assessment of the views of low-income US mothers. J Reprod Infant Psychol 2016; 34:35–48Crossref, MedlineGoogle Scholar

25 Patton MQ: Qualitative Research and Evaluation Methods, 2nd ed. Thousand Oaks, CA, Sage, 2001Google Scholar

26 Derogatis L: SCL-90-R Administration, Scoring and Procedures Manual II for the Revised Version. Towson, MD, Clinical Psychometric Research, 1983Google Scholar

27 Cooper C, O’Cathain A, Hind D, et al.: Conducting qualitative research within clinical trials units: avoiding potential pitfalls. Contemp Clin Trials 2014; 38:338–343Crossref, MedlineGoogle Scholar

28 Dedoose Version 6.2.7, Web Application for Managing, Analyzing, and Presenting Qualitative and Mixed Method Research Data. Los Angeles, SocioCultural Research Consultants, 2015Google Scholar

29 Spitzer RL, Williams JB, Kroenke K, et al.: Validity and utility of the PRIME-MD patient health questionnaire in assessment of 3000 obstetric-gynecologic patients: the PRIME-MD Patient Health Questionnaire Obstetrics-Gynecology Study. Am J Obstet Gynecol 2000; 183:759–769Crossref, MedlineGoogle Scholar

30 Sheehan DV, Lecrubier Y, Sheehan KH, et al.: The Mini-International Neuropsychiatric Interview (MINI): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59(suppl 20):22–33MedlineGoogle Scholar

31 Blanchard EB, Jones-Alexander J, Buckley TC, et al.: Psychometric properties of the PTSD Checklist (PCL). Behav Res Ther 1996; 34:669–673Crossref, MedlineGoogle Scholar

32 Caruso DR, Salovey P: Coaching for emotional intelligence: MSCEIT; in Psychometrics in Coaching: Using Psychological and Psychometric Tools for Development. Edited by Passmore J. London, Kogan Page, 2008Google Scholar

33 Mayer JD, Salovey P, Caruso DR: Emotional intelligence: new ability or eclectic traits? Am Psychol 2008; 63:503–517Crossref, MedlineGoogle Scholar

34 Lizerett NP, Oberst U, Chamarro A, et al.: Evaluation of emotional intelligence in patients with psychopathology: preliminary results using TMSS-24 and the MSCEIT [in Spanish]. Ansiedad y Estres 2006; 12:355–364Google Scholar

35 Brackett MA, Rivers SE, Shiffman S, et al.: Relating emotional abilities to social functioning: a comparison of self-report and performance measures of emotional intelligence. J Pers Soc Psychol 2006; 91:780–795Crossref, MedlineGoogle Scholar