A Qualitative Assessment of the Parenting Challenges and Treatment Needs of Mothers with Borderline Personality Disorder

Abstract

There are currently no empirically supported interventions to target parenting among mothers who have Borderline Personality Disorder (BPD). The current study uses Consensus Qualitative Research (CQR) methodology to: I) learn about mothers’ experiences of parenting with BPD, and II) identify treatment modifications to Dialectical Behavior Therapy (DBT) as suggested by mothers with BPD who are currently engaged in DBT skills training. Twenty-three mothers were recruited from intensive outpatient and partial hospitalization programs that teach DBT skills. A total of 9 focus groups that met one time were conducted asking women a series of questions regarding their experiences of parenting with BPD and how they would modify DBT to address parenting issues. Using the CQR approach, we coded domains and categories that were discussed by mothers in the focus groups. Coding revealed that mothers with BPD wished parenting was integrated more in their current DBT skills groups. In addition, one of the most prominent themes to emerge was that parenting is particularly stressful to mothers with BPD and is associated with guilt, uncertainty, and worry. Finally, mothers offered many ideas for how to integrate parenting-focused interventions into DBT. The CQR method revealed gaps in current treatment for mothers with BPD and provided useful ideas for how to modify DBT to target parenting and integrate these modifications into other approaches for treating mothers with BPD.

Empirical studies on maternal borderline personality disorder (BPD) are disturbingly scarce given that many mothers are attempting to parent with this debilitating mental health condition and many children are being raised in the context of serious psychopathology (Stepp, Whalen, Pilkonis, Hipwell, & Levine, 2012). Across development, maternal BPD, as well as the resulting at-risk home environment (Fruzzetti, 2012), are associated with negative child developmental outcomes, such as insecure attachment status and emotion dysregulation (Crawford, Cohen, Chen, Anglin, & Ehrensaft, 2009; Hobson, Patrick, Crandell, García-Pérez, & Lee, 2005; Macfie, 2009b). Evidence also suggests that the combination of maternal personality disorder with depression imparts greater risk for negative child outcomes than does maternal depression alone. For example, maternal personality disorder interacts with maternal depression to predict higher levels of dysregulated infant behavior (Conroy et al., 2012). In addition, older children of mothers with both BPD and depression show particularly high risk for developing depression themselves (Abela, Skitch, Auerbach, & Adams, 2005).

Despite the implications of parenting with BPD and the effects of being raised by a BPD mother, clinical recommendations for treating a woman with borderline personality disorder who has children have not been well articulated (Stepp et al., 2012). Although there are numerous effective treatments for BPD (Leichsenring, Leibing, Kruse, New, & Leweke, 2011), and many effective parenting treatments (Cedar & Levant, 1990; de Graaf, Speetjens, Smit, de Wolff, & Tavecchio, 2008; MacKenzie, Fite, & Bates, 2004), these two areas of intervention have not been substantially integrated. This presents a serious problem in which a) the mental health needs of women with BPD may not be addressed in standard parenting interventions, and b) standard mental health treatment for BPD typically does not address parenting problems. Although not all mothers with BPD will have children with mental health issues, many will face challenges around the role of parenting that treatments to date do not address. Focus groups of mothers with BPD who were in psychiatric treatment were interviewed to begin to elucidate some of these experiences of being a mother with BPD in treatment.

Dynamic and behavioral theoretical models of parenting interventions have considered whether maternal BPD, or associated phenomena such as mothers’ own insecure/disorganized attachment, necessitates modifications to the treatment approach. Although an exhaustive review of the modifications to parenting interventions is provided elsewhere (see Stepp et al., 2012 for a review on current treatment approaches), we will briefly review some key work in this area. Dynamic theories focus on attachment interventions which assist the mother in exploring how her own attachment style influences her interactions with her child, as transmission of disorganized attachment status has been noted across generations (Lyons-Ruth, Yellin, Melnick, & Atwood, 2003). Such intervention models, which have been used with at-risk populations, include Watch, Wait, and Wonder (Cohen et al., 1999), Preschooler-Parent-Psychotherapy (Cicchetti, Rogosch, & Toth, 2000), and Circle of Security (Marvin, Cooper, Hoffman, & Powell, 2002). Of the attachment interventions, only Watch, Wait, and Wonder, an approach that enhances maternal sensitivity to infant and toddler cues, has been used in its original form with 20 mothers with BPD (Newman & Stevenson, 2008). However, the study did not include an evaluation of its effectiveness across participants.

From a behavioral viewpoint on parenting interventions, a variety of parent management training programs have been established as evidence based (Chambless & Hollon, 1998). However, a recent paper (Maliken & Katz, 2013), reviewed how parental emotion dysregulation, a transdiagnostic feature of several forms of psychopathology and a cardinal symptom of BPD, mitigated the effectiveness of parent management training (PMT) interventions. The authors cited recruitment, retention, engagement, skills acquisition, and skills enactment as key barriers by which treatment results were less successful for these families. Recognizing these issues, enhanced versions of parenting interventions have been developed to target key symptoms of psychopathology, such as emotion dysregulation, that interfere with parenting (Sanders et al., 2004; Webster-Stratton & Reid, 2012). Despite these augmentations, it remains unclear whether modifying existing parenting treatments are adequate for addressing the needs of mothers with BPD.

Perhaps a key reason that parent management training does not adequately address parental psychopathology is that these treatments were originally designed to treat child behavior problems (Patterson, Chamberlain, & Reid, 1982) and only targeted parental psychopathology as a method to improve learning the intervention material. When considering the adaptations of attachment based and parent management training interventions with mothers with BPD or associated features, none have reached the point of integration to be considered a viable treatment option to date. Although new interventions are being developed to target dually-disordered mother-child dyads, (Chronis-Tuscano & Clarke, 2008; Swartz et al., 2008), these treatments focus on maternal depression, with no treatments targeting maternal BPD. Findings from basic science studies suggest that maternal BPD may be associated with certain aspects of parenting (Zalewski et al., 2014), including the increased used of psychological control. Further, small studies have reported that mothers with BPD are less responsive and more intrusively insensitive to their infants (Crandell, Patrick, & Hobson, 2003), and less sensitive and structuring in their interactions with their infants (Newman, Stevenson, & Boyce, 2007). Another study found that mothers with BPD, as compared to mothers with depression or no disorder, demonstrated more disrupted affective communication and more frightened and disoriented behavior in response to their distressed infants’ bids for attachment (Hobson, Patrick, Hobson, Crandell, Bronfman, & Lyons-Ruth, 2009). Additionally, BPD compared to other disorders, such as maternal depression, may vary in regards to other practical matters that also impact the parenting role. For example, as individuals with BPD are overrepresented within inpatient settings (Bender & Skodol, 2007), these mothers will more likely have to make considerations around child care or provide explanations to family members regarding hospitalization, compared to mothers with other forms of psychopathology. These practical concerns are often not addressed as part of standard parenting interventions. Thus, maternal BPD may necessitate targeted treatment development efforts.

The limitations inherent with targeting psychopathology within parenting interventions calls for an alternative approach. One such alternative is integrating parenting-focused interventions with existing adult psychiatric treatment approaches for women with BPD. In this way, maternal role functioning (Logsdon, Wisner, Sit, Luther, & Wisniewski, 2011) could be viewed as an treatment target within standard adult treatment for BPD, encouraging treatment of the whole individual and not just presenting symptoms. To date, there is no empirical data testing the effectiveness of integrating parenting interventions with existing adult psychiatric treatment for BPD. Therefore, the present study uses qualitative methodology to understand the experiences of being a mother with BPD who is in psychiatric treatment and assesses women’s preferences for integrating interventions focused on the role of parenting into their current treatment. Below, we review the literature on parenting with BPD and the current treatment approaches for treating BPD.

Experiences of Parenting with BPD

Although there are few studies examining the role of parenting from the perspective of individuals with BPD, one study found that mothers with BPD reported being more distressed, less satisfied, and less competent in their parenting roles (Newman et al., 2007). Considerably more empirical work has examined parenting experiences in those with other forms of psychopathology. One such study employed qualitative methods to characterize the experiences of parenting with mental illness (Nicholson, Sweeney, & Geller, 1998). Primarily sampled from parents with affective disorder diagnoses and treating clinicians, the researchers identified 4 themes from the parents’ experiences. These parents talked about the stigma of mental illness, day-to-day stresses of parenting, managing mental illness, and issues around custody and contact with children. One of the overarching observations from this qualitative study was that mothers had insight and awareness that their mental illness impacted their parenting and their children. Thus, they were able to reflect on their own behavior and mental states and their influence on others, which is closely akin to the capacity for mentalization (Fonagy, 1991). Fonagy and colleagues (e.g., Bateman & Fonagy, 2004; Fonagy, Gergeley, Jurist, & Target, 2002) have written extensively on how this capacity is often impaired among individuals with BPD, particularly in the context of high affective arousal. Recent evidence suggests that individuals with BPD show a tendency toward hypermentalization, i.e., the overinterpretation or overattribution of overly complex intentions or mental states, leading to inaccuracies and confusion between the mental states of self and others (Sharp et al., 2011, 2013). However, it remains unknown whether mothers with BPD show mentalizing failures with respect to their insight and perspective-taking on their own parenting or relationships with their children.

Current Treatment Approaches for Mothers with BPD

One of the most widely studied of several efficacious treatments for BPD is Dialectical Behavior Therapy (Linehan, 1993a), a comprehensively integrative form of psychotherapy rooted in dialectical philosophy (Heard & Linehan, 1994) and originally developed for women with BPD who engage in non-suicidal self-injury. DBT synthesizes the principles of Zen practice with behavioral principles of change, and also draws from various therapeutic perspectives. Similar to other empirically supported approaches for treating BPD such as transference focused psychotherapy (TFP; Kernberg, 1984), DBT is a structured treatment that follows a hierarchy of treatment targets, provides high levels of therapist support (e.g., consultation groups), and emphasizes non-judgmental stance and validation balanced with encouragement for change (Swenson, 1989). One of four treatment components in DBT is Skills Training that covers four modules: Mindfulness, Distress Tolerance, Emotion Regulation, and Interpersonal Effectiveness. DBT skills are shown to be a critical part of increasing the use of effective and skillful behaviors in clinical populations (Neacsiu, Rizvi, & Linehan, 2010; Stepp, Epler, Jahng, & Trull, 2008). DBT treatment as well as skills training are often modified to address the needs of populations that DBT has been adapted to, such as children (Perepletchikova, Ansell, & Axelrod, 2011), adolescents (Miller, Rathus, & Linehan, 2007), and those with eating disorders (Safer, Robinson, & Jo, 2010).

It should be noted that although there have been attempts to teach parents DBT skills, the parents were not mothers with BPD. The treatment approaches that teach parents DBT skills include: 1) Adolescent DBT in which parents and teenagers concurrently learn the skills (Miller et al., 2007); 2) clinical case studies for parents with difficulties regulating their affect (Ben-Porath, 2010); 3) and a self-help book written on how to use DBT skills for parents of children with intense emotions (LCSW-C & LICSW, 2009). Unfortunately, none of these approaches have brought parenting skills to DBT treatment for mothers with BPD. In a review paper aimed at identifying parenting behaviors associated with parental BPD, Stepp and colleagues outlined several treatment development ideas (Stepp et al., 2012). These included teaching parents about child development, teaching parents skills to promote consistency and warmth, and teaching parents to incorporate mindfulness into their parenting. These ideas were drawn from the scant literature on parental BPD but do not include treatment preferences articulated by BPD women themselves.

Qualitative Research

The purpose of this qualitative study was to learn about the experience of being a mother with BPD currently in treatment, with the ultimate goal of gathering information that can be utilized in developing a treatment that integrates parenting-based interventions into existing treatments for BPD. Although in this study we focused on women in DBT-informed treatment, we believe that these results can be generalized to other efficacious treatments for BPD that share several common factors or putative mechanisms of change (Bliss & McCardle, 2013; Swenson, 1989; Weinberg et al., 2011). We followed the steps depicted in the Consensual Qualitative Research (CQR) design (Hill, Thompson, & Williams, 1997). We choose this qualitative approach due to its transparency of methodology, which can often be elusive in qualitative designs (Webb & Kevern, 2001), because of its similarity to other well known qualititaive methodologies, and because the CQR theory fits well with our study goals. Specifically, CQR methodology is most similar to the grounded theory qualitative approach, which seeks to develop conceptual links between related concepts about a phenomenon, rather than subscribing to the belief of a pre-existing reality (Glaser, 2012). A major methodological difference between CQR and grounded theory is that CQR uses the same protocol to collect all data, rather than continually adjusting the protocol based on information gathered from participants. Because we targeted a very specific sample at the initiation of the study---mothers with BPD who were in DBT-informed treatment, CQR was most appropriate as it was not necessary to iteratively go between data collection and data analysis.

The CQR method encourages researchers to evaluate their own theoretical orientation and biases prior to pursuing qualitative work, which is described here. The authors of this research have been trained in DBT to a varying extent. DBT therapists are trained to monitor their own judgmental thoughts while also recognizing that therapists will have judgmental thoughts and reactions to their clients at some point during treatment. Being trained under this assumption, we believe that this is well-suited for our exploratory approach. We wish to understand their viewpoints, while not “sugar coating” the challenges these mothers may exhibit in parenting. However, as with the fundamental dialectic of acceptance and change (Linehan, 1993a), we believe that we cannot work to change behavior unless we have actually accepted what is happening in reality. We believe that the qualitative nature of this work, asking the actual women for their experiences, serves as a synthesis.

Method

Procedures for Collecting Data

Recruitment

A total of 23 women were recruited from an intensive outpatient program (meeting approximately 2–3 times per week) or partial hospitalization program (meeting 5 days a week) in Western Pennsylvania. The intensive outpatient and partial hospitalization program would best be described as a DBT-informed approach, including DBT skills training, individual therapy, and weekly psychiatry appointments. The individual psychotherapy would unlikely meet fidelity criteria and unlike standard DBT, phone coaching was not available. Women diagnosed with BPD at the hospital used for recruiting would most likely receive treatment from these programs. Anyone who was in the outpatient or partial hospitalization treatment and who was a mother was invited to participate over a 4-month window. Recruitment and groups were conducted at a bi-weekly rate, to permit a turn over in patient samples. Five minutes prior to the start of their group, we recruited potential participants by announcing to the entire group room. Potential participants were told that we were interested in learning about their experiences of being a mother with BPD who was receiving psychiatric treatment. Women could be at any point in their treatment to be eligible for participation. Interested mothers typically stayed after their group therapy time for an additional 1–2.5 hours to be interviewed in focus groups (ranging from 1–8 women participants) with women they already knew from treatment. Despite the range of women in each group, interviewers ensured that each woman had a chance to respond to each prompt.

The current study data were de-identified and all procedures were approved by the Internal Review Board of the University of Pittsburgh. The interviewer read the informed consent and obtained verbal consent from all women. Women completed three questionnaires and were audiotaped during the interview process (Interview questions listed in Appendix A). Women were paid for their time. After all the interviews were completed, they were transcribed by the Qualitative Data Analysis Program, University Center for Social and Urban Research, University of Pittsburgh. To protect anonymity, women’s names were not transcribed.

Participants

On average, mothers’ were 39 years old (SD=14; Range =19–64). Of the 23 mother participants, 8 women identified as Black, 12 as White, 1 as multiracial, 1 filled in her own race, and 1 woman declined this question. Eleven women were married or in a committed relationship, 3 were single and had never been married, 3 were separated and 6 were divorced. Ten of the women had completed college or a graduate degree. Ten women reported having a household income less than $30,000 a year, 10 women reported their household income as between $30,000–$75,000 and the remaining 3 women reported earning over $75,000 a year. Half of the sample was employed in some capacity. Twelve women had 1 child, 4 had 2 children, 3 had 4 children, and 4 women had 5 children. Children’s ages ranged from 9 months to 45 years old.

Measures

The second author developed an 8-question interview, conducted with all groups. The interview questions focused on the experiences associated with being a mother who has BPD and mothers’ treatment preferences. Example prompts included, “How do you communicate with your children about Borderline Personality Disorder?” and, “How would you change or improve the current DBT program to focus on parenting issues and your parent-child relationship?”

In addition, we collected two measures to help characterize the sample in regards to BPD symptoms and parenting stress. Women completed the Personality Assessment Inventory – Borderline Features Scale (PAI-BOR; Morey, 1991), a 24- item self-report measure with choice options ranging from 0-‘False, Not at all True’ to 3-‘Very True’, such that higher scores indicate greater severity. The average score was 42.09 (SD = 13.51, Range = 19–63; α = .90), with approximately 61% of the sample scoring 38 or higher, the clinical cut-off score, suggesting the majority of women experienced clinically significant BPD symptoms (Trull, 1995). The PAI-BOR has four subscales, Affective Instability, Negative Relationships, Identity Problems, and Self-harm, of which 78.3% of women continued to meet the clinical cut-off on at least one subscale, despite being engaged in intensive outpatient psychiatric treatment. Women also completed the Parenting Stress Scale (Berry & Jones, 1995), an 18-item measure with 5 choice options, ranging from 1-‘Strongly Disagree’ to 5-‘Strongly Agree’. The average score was 47.61 (SD=13.97), with higher scores indicating higher levels of parenting stress. In the original article, an average score of 37.1 (SD=8.1) was reported for mothers of control children, suggesting that mothers with BPD reported over 1 standard deviation greater of parenting stress than the original sample.

Analysis Team

The 4 authors each served on the coding team, including a faculty member in psychiatry, a post-doctoral researcher in psychiatry, a clinical psychology intern, and a psychology graduate student. At the time of data collection, the intern conducted 8 out of the 9 focus groups, and the graduate student team member was present at 4 of the focus groups. The faculty member conducted 1 focus group, which the graduate student also attended. At the time of coding, the clinical psychology intern was a post-doctoral researcher in the department of psychiatry. The fourth author served as the auditor.

The analysis team researched the CQR method and trained by reading the core articles written on the how to conduct CQR (Hill et al., 1997; 2005). As was recommended by these articles for training purposes, we also read exemplar articles using this approach (Knox, Hess, Williams, & Hill, 2003), and one article in which the developer of the methodology was not on the team (Wettersten et al., 2004).

Procedures for Analyzing Data

Identification of domains and core ideas

Each of the primary team members read one of the transcripts to identify domains (i.e. topic areas) and core ideas (i.e. abstracts or brief summaries). An initial list of domains and core ideas was drafted, and the analysis team met to discuss. The core domains changed throughout the remaining coding process as team members reached agreement on better ways to describe the domains.

Cross analysis

Next, the team members proceeded to create categories underneath the domain topics. The team members also coded each transcript to indicate whether they noted the category as present within a particular transcript. The team members then compared ratings of whether they thought the category was discussed and disagreements served as effective means to recognize when a category was too broad (i.e., needed to be broken into more categories) or too specific (i.e., needed to be grouped with another category). Keeping with the labels often used in the CQR methods, categories were considered general if they applied to all groups, typical if they applied to at least half of the groups (5–8 groups), and variant if they applied to fewer than one half but at least 2 groups (2–4).

After the initial identification of domains and coding of categories, the auditor reviewed and provided feedback at both the level of the domain and category coding. The auditor read through a subsample of transcripts while reviewing the domains and categories. The auditor also read through notes made on the first author’s transcripts that highlighted examples from the transcript that were coded into categories. The auditor made several suggestions that were then discussed by the primary team. Some examples of changes made by the auditor that were employed in domain or category revisions will be described in the results section.

Results

Because the interview had a natural shift from discussing the experiences of being a mother with BPD to discussing treatment development ideas, the results are divided into two segments. The domain names, categories, and labels are listed in Tables 1 and 2 for the experiences and treatment development discussions, respectively.

Mothers’ Experiences with Parenting

Concerns with talking to child about mental illness

One domain that typically emerged regarded BPD mothers’ experience of whether or not to tell their child about their diagnosis. Two categories emerged: (1) developmental concerns and (2) not knowing how to communicate about disorder.

Developmental concerns

The category of developmental concerns was typically discussed. The mothers discussed choosing not to tell their child about their diagnosis because their child was too young to understand or because they were concerned the information would scare their child in some way or negatively alter the child’s sense of their mother. A mother stated:

It also becomes a worry for kids sometimes, you know? Like, genetically. Like, “Oh, am I gonna have the same problems Mom does?” Like, my mom’s schizophrenic, so I wonder all the time, “Am I gonna be schizophrenic when I grow up?” Just like anything, like diabetes, you know? Like, it’s run in my family, I’m gonna have diabetes. You know, like, you’d start freaking out all the time.

Not knowing how to communicate about the disorder

This category was typically discussed. Although at times a mother’s description of not knowing how to communicate about her disorder to her child overlapped conceptually with developmental concerns (i.e., she was not sure how to explain her crying to a very young child), as a group, we agreed that there was enough conceptual distinction between these categories to warrant separation. Mother’s of children of all ages described their difficulty with knowing how to explain their diagnosis or why they were in treatment to their children. This included mothers who have attempted to explain their diagnosis and felt uncertain about how it went, as well as mothers who were hesitant to bring up the topic to their children because of uncertainty about what type of or how much information should be discussed.

During our coding meetings, team members shared some of their own thoughts on this topic. The first and second author, having worked clinically in the setting sampled in this paper, noted the possibility that the mothers’ challenge with communicating about their disorder with their children may stem in part from their own lack of knowledge about their disorder. Despite these women being in psychiatric treatment, many appeared not to know all of the symptoms that comprise their diagnosis. In fact, as was stated earlier, when recruiting our sample, some women were uncertain if they had BPD, at times telling the interviewer that this diagnosis sounded familiar to them and they thought they had “it.” Therefore, although not coded as a category, because this would not have been described in the transcripts, our clinical experience working with BPD mothers informs our opinion that some mothers may struggle to communicate about their disorder because they are not entirely sure what their diagnosis is or what it means.

Impact of BPD on children

Many interview questions targeted mothers’ thoughts and beliefs about whether or how BPD affected their children. Four categories emerged: (1) shared traits, (2) role changes/reversal, (3) consequential stressors, and (4) mother’s emotion dysregulation.

Shared traits

Shared traits were typically discussed as mothers noted several examples of ways they saw their children exhibiting BPD-like traits. Mothers noted observing impulsivity, anger/aggression, moodiness, and sadness in their children. Typically, mothers spoke about these shared traits either with a tone of empathy/understanding for their child (“She’s very impulsive, just like I am, too”) or worry. As an example of a mother being worried or concerned about her child’s similar traits, a mother shared:

They picked on her so bad that she just, like, went off on them. And I’m like, “What do you mean ‘went off’?” She’s like, “I just cussed them all out, screamed, and told them to back the H away from me.” But I realized that I’ve been that upset and I’ve reacted that way, and my daughter has seen me, at times, react that way.

When discussing shared traits, many mothers expressed both guilt and empathy for their child. Despite BPD women often being viewed in a deficit model of parenting (Nicholson et al., 1998), it is possible that these mothers could have more empathy and less judgment toward some of their children’s behaviors, particularly the traits they believe their children ‘inherited’ from them.

Role changes/reversal

For this category, the team originally coded the data into ‘emotional protection’ and ‘child managing mother’s illness.’ However, we ultimately decided the category of ‘Role changes/reversal’ fit better because the team, at times, struggled to accurately code between ‘emotional protection’ and ‘child managing mother’s illness’, as was the case with the example below. As the original category descriptions imply, mothers described instances in which their children comforted them as well as managed aspects of the mother’s care. In an example that highlights both emotional protection and managing a mother’s illness, a mother with an older daughter shared how her daughter stayed home with her and said:

But she is aware of my self-harm now, and she—that bothers her. She checks, you know—

Interviewer: Oh, she does?

Yeah, just my arms. You know, she’ll make sure that she’s looking at my arms all the time.

Taken in the context of the transcript, this mother described her daugther’s behavior stemming from her fears about her mother’s safety, as well as hinting that her daughter would take action when her mother self-harmed. Throughout the interviews, there were several mothers who described instances in which the child was behaving more like the adult or parent in the dyad.

Associated life stressors

A variant category was ways in which having BPD affected other aspects of their life, which mothers’ felt had negatively impacted their children. In most all cases, this referred to the quality of their marriage or their inability to maintain employment.

Mothers’ emotion dysregulation

A typically discussed category was mothers’ description of how their own emotion dysregulation negatively impacted their child/children. Emotion dysregulation is a hallmark feature of BPD so, even though not explicitly elicited from the women, many were aware that their struggles with emotion dysregulation affected their parenting. The example provided by one woman was a prototypical description:

And some of it isn’t hard at all, but the emotional part, like, you know, with having ups and downs in my mood is really hard. You know, like, just wanting to – like, it’s not that I don’t want to be with my kid, but I just don’t want to do anything, you know, and then he takes it as, “I must have done something wrong” or “I must have made Mom upset,” you know, “I’ve done something to make her feel this way.” And it has nothing to do with him. It’s just I just have really bad mood swings and anger.

From these common examples of emotionally dysregulated expressions, our team also commented that many women did not know how to recover or reconcile with their child afterwards.

Parenting challenges

The following four categories emerged primarily after mothers were prompted to answer questions regarding what challenges they observed as specific to having BPD: (1) custody issues, (2) guilt/worry/uncertainty about parenting role, (3) transgenerational reflections, and (4) burden of parenting. However, there were several instances throughout the interview where mothers spontaneously discussed challenges they had with parenting, most often in regards to the guilt/worry/uncertainty they had around their role as mothers.

Custody issues

In 2 groups, there was at least one mother who shared that her child was either temporarily removed from her home or was in the care of another family member.

Guilt/worry/uncertainty

As stated, the most prominent category discussed in all 9 groups was mothers’ guilt/worry/uncertainty about her parenting. This category describes mothers’ lack of confidence around her parenting decisions, guilt over past transgressions, and worry about how her children will adjust in life. This category was originally divided into 3 categories. However, many of the mothers’ descriptions encompassed both guilt and uncertainty, both guilt and worry, or in some cases all three categories. This category spanned a vast range of examples, including uncertainty if a mother should let her child play with a particular toy to uncertainty about retaining her role as a parent. The following example from a mother in group 5 illustrates a combination of guilty, worry, and uncertainty:

I tell myself that I don’t want to be a parent, but I know that that’s not right. I do, I want to be a parent, I want to be a successful parent, I want to be a happy parent, but it’s like, it’s a fantasy, [voice stars to crack] and that those things don’t exist…. I need to escape now while he’s still young before he’ll have memories of me walking out.

From this example and in a few other groups, team member/author 3 noted a pattern in which you could hear mothers grappling with the uncertainty aloud. The mothers would vacillate between stating something negative about parenting to then stating that they loved their child.

Reflections on own childhood

Another variant theme was when mothers were describing their parenting challenges, a few remarked on how they have had some issues since childhood that they believe were a result of their own upbringing. Examples of these reflections include noting the parallels between how their mothers related to men with how they themselves relate to men, abandonment issues, or even their fear of labeling their child’s mental health issues because they had bad experiences with being labeled with issues when they were children.

Burden of parenting

Mothers typically noted that parenting felt like a burden or eluded that the role was stressful (“When he was an infant there was not any trouble, but when he start to speak out – it became a nightmare.”). When coding the transcripts, the team members all noted that there were very few instances in which mothers commented on the joys of parenting. Although the interview questions realistically pulled more for negative responses to parenting, we imagine that if the same questions were asked of a group of mothers without psychopathology, the valence of the responses would have been more mixed. Similar to the category of burden of parenting, the auditor noted observing instances where mothers appeared to struggle with managing closeness versus distance with their children, and that overall, this challenge for mothers felt like an additional source of stress. Some mothers described their children as feeling ‘clingy.’

Parent-Child DBT

The focus group questions shifted to discussing mothers’ perspectives on participating in a DBT program that focused on parents and children called Parent-Child DBT (PC-DBT). The program is currently being developed and the focus group served as a method to learn about mothers’ preferences for treatment. The following domains and categories largely emerged from this portion of the interview.

Positive aspects of PC-DBT

When asked about the benefits or difficulties of participating in a PC-DBT program, mothers were very positive about the idea. We coded mothers’ responses into three primary categories of positive aspects of participation: (1) commonality with other mothers, (2) fear of judgment of own parenting, and (3) fear of exposing children to BPD-traits in other mothers.

Commonality with other mothers

When asked specifically about what the benefits to participating in PC-DBT would be, mothers typically noted that being in a group with mothers would be helpful. One mother commented:

But a lot of the people in here don’t have kids, so it’s mainly about them and their issues and… when you’ve got kids, sometimes you need to talk about your kids. Sometimes it’s what your kids are doing that are getting you to these points, you know? And a lot of parents don’t like to admit it, but their kids do things that really get them going….but having issues like this makes you feel like you are the one causing the problems…

It was the first author’s observation that many of these mothers would have preferred being in a group with all mothers so that topics such as parenting would be focused on explicitly.

Children could benefit from learning skills

Mothers typically commented that they would like their children or adolescents to learn DBT skills. Mothers appeared to make these comments both when their children were doing well and when they perceived their child as having anger or impulsivity problems as well. A mother stated:

…if the kids were in the group, I think it would teach them how to deal with issues they were fighting with other school kids ….And I think that that would help the kids if they learned DEAR MAN¹…

Increase confidence as a parent

Similar to the theme expressed above about worry and uncertainly in the parenting role, mothers typically commented that they thought PC-DBT would help decrease their self-doubt. We noticed a range in mothers’ ability to articulate specifics of how DBT could increase their confidence in parenting. Some mothers were able to provide specific examples of using DBT skills in the context of parenting while other mothers felt it could help but weren’t able to articulate specifics. The example below demonstrates how one mother views PC-DBT as a treatment that could help her negotiate her own uncertainty as a parent. This mother expressed:

My biggest problem is self-confidence and identity, like identifying what kind of a mother I want to be. A lot of my problems are essentially “who do I want to be as a parent. Do I want to be my child’s best friend, or do I want to be a disciplinarian”?

This example is one of many in which women are struggling to understand both themselves and their role as parents, thus leading to a reduced sense of confidence in the parenting role.

Potential barriers and clinical considerations with PC-DBT

Similar to the Positive Aspects of PC-DBT section, mothers described what they could imagine potentially getting in the way of their participation. As noted by our auditor, many mothers noted concerns that would hamper their desire to participate, however, clinicians or program coordinators could make adjustments that would allay such concerns. Therefore, we chose the domain title to encompass both barriers and considerations, rather than split these into two domains.

Fear of judgment of own parenting

Most often in response to our query of what might be difficult about participating in PC-DBT, mothers typically mentioned fear of being judged as a parent. Some mothers even mentioned fear that a child protection agency could be notified if some of their parenting behaviors were taken out of context. Interestingly, one mother even suggested that PC-DBT start with radical acceptance, in that mothers would have to recognize for themselves and to each other that they have engaged in parenting behaviors that might elicit shame.

Fear of exposing kids to other BPD parents

Some groups discussed the pros and cons of having children attend skills group with their mothers. During these discussions, a variant concern mentioned by a few mothers was that of how other women’s behavior may affect their children. Mothers expressed concern about having their child witness emotionally driven behavior of other women.

Group structure

Mothers were asked directly about how they would design PC-DBT. Mothers in every group talked mainly about how they envisioned the group would be structured. Although we originally had separate categories for the different types of group structures, mothers offered so many variations on group structure ideas that we opted to describe some of the ideas here rather than divide the category further. Mothers mentioned having the children in the same group as them, having separate parent and child groups with some mothers wanting their children to be entertained/taken are of, while other mothers wanted their children to learn child DBT. Other mothers expressed a desire to have several weeks in which they were learning distress tolerance, after which the children would be brought in for therapy as well. Other suggestions included a therapist working individually with the mother and child, watching the mother interact with her child and providing feedback, or an individual therapist working with the child directly. It was our observation that mothers’ suggestions were influenced largely by: a) her child’s age, b) whether her child had mental health issues, c) how well the mother and child got along and whether the mother thought her child would want to attend, and d) if the child knew about the mother’s mental health issues and treatment engagement.

Timing

Mothers most often noted that timing of the group would be related to the group structure employed. For instance, mothers who envisioned PC-DBT including school-aged children as part of the treatment noted that the group would need to meet after school or on the weekends.

Logistical concerns

Mothers typically voiced several concerns regarding the logistics of PC-DBT, as this was prompted by our focus group interview questions. Mothers expressed concern about the location of the treatment, as well as noting that if treatment were in the evening, they would need help providing dinner. In addition, mothers expressed concern about cost and insurance coverage of treatment.

Discussion

This paper explored the potential of targeting parenting within an established adult psychiatric treatment for women with BPD as an alternative approach to targeting parental psychopathology within established parenting interventions. Although providers would not be prohibited from targeting parenting in standard care, to date, there are no empirical studies testing such a model for women with BPD. Given the paucity of research in this area, a qualitative research study was initiated. The purpose of this research study was to: I) understand the experiences of parenting with BPD, and II) consider treatment modifications to DBT treatment as suggested by mothers with BPD who were currently engaged in DBT skills training. The following sections are organized by these two main study purposes and then by domain levels identified by our qualitative methods. The study findings are discussed in the context of future treatment development efforts that aim to integrate knowledge between adult psychiatric treatment and parenting interventions.

Experiences of Parenting with BPD

Although limited, there is increasing evidence that maternal BPD increases offspring risk for developing psychopathology through heritability and parenting practices (Macfie, 2009a; Stepp et al., 2012; White, Gunderson, Zanarini, & Hudson, 2003). However, there is little research to date describing the actual experiences that mothers with BPD face when parenting from their own perspective. The following sections review the domains identified while keeping a focus on treatment development.

Concerns with telling child

Mothers discussed not knowing developmentally appropriate ways to reveal having a BPD diagnosis to their children. It is possible that mothers with BPD struggle with this because they do not understand their own diagnosis. In fact, it has been shown that a BPD diagnosis is not always disclosed to the patient, which means that a patient may not have much information regarding their diagnosis or it is not adequately explained (Lequesne & Hersh, 2004). There are few resources that exist for providing parents with guidance on whether or how to tell their children about their mental illness. One such developed resource is through a website, Children of Parents with a Mental Illness (http://www.copmi.net.au/) that offers developmentally focused ideas for informing children about parental psychopathology. These handouts appear to focus on normalizing the experience and assuring that children will be cared for. Despite the soundness of this clinical opinion, it remains unclear if such advice pertains to mothers with BPD. For instance, some mothers in our group revealed that their teenage daughters were aware of their self-harm. In instances in which parents’ illness is severe and not well managed, it might be ill-advised to tell their child that everything okay, when in reality, a mother may be seriously considering ending her own life. Attempting to provide a child reassurance when a mother is not in a position to do so could actually be harmful. More consideration is needed on this topic of disclosure, normalization of parental mental illness, and the clinical advice that parental reassurance should be provided to a child. Future research intervention efforts should focus on helping mothers identify the pros and cons, a DBT skill, regarding whether they should disclose aspects of their mental health issues to their children.

Impact of BPD on children

Similar to a qualitative study conducted on mothers with mental illness (Nicholson et al., 1998), mothers displayed considerable insight into how their children were affected by the mothers’ BPD diagnosis. Many mothers were aware that their own children exhibited similar traits such as angry outbursts or impulsivity. In some cases, mothers even noted that some of their children had similar traits whereas their other children did not, suggesting that for at least these mothers, it is unlikely that they were ‘overpathologizing’ their child’s behaviors, as has been raised as a concern in the maternal depression literature (Biederman, Mick, & Faraone, 1998; Chilcoat & Breslau, 1998). Clinicians working with mothers with BPD may use the empathy these mothers have for their children as a treatment strength. Additionally, mothers noted that their symptoms of BPD also impacted other aspects of their life, such as the quality of their marital relationships, which in turn negatively impacted their children. Self-awareness was also evident when mothers were able to cite specific instances when their emotion dysregulation hurt the quality of their relationship with their child. Parental emotion dysregulation affects the way in which parents respond to their child’s emotions and behaviors (Bariola, Gullone, & Hughes, 2011), with one study demonstrating that parental emotion dysregulation was one of the more important components of BPD that related to negative parenting behaviors (reference withheld for blind review; provisionally accepted).

These observations suggest that mothers with BPD may have some capacity to reflect on their own behavior and their influence on the experiences of their children, suggesting at least a moderate degree of capacity to mentalize (e.g., Fonagy, 1991) with regard to parenting. The level of insight that these mothers demonstrated into how their behavior and illness affects their children may have positive prognostic value for the success of parenting-focused interventions, and will shape the development of intervention efforts. For instance, if mothers with BPD did not demonstrate awareness regarding how their diagnosis impacted their relationship with their children, early intervention efforts would need to spend considerable time working with mothers on seeing connections between their BPD diagnosis and behaviors that impact their children. This capacity to reflect on their own and others’ behaviors in terms of underlying mental states is the core mechanism targeted by mentalization-based therapy (Bateman & Fonagy, 2004). Although empirical studies are needed to determine whether mothers with BPD show failures or impairments in this capacity with respect to their interactions with their children, the level of insight they demonstrated in this qualitative study with regard to the effects of their own behaviors on their children is encouraging. Acceptance based skills, such as radical acceptance (Linehan, 1993b), as well as mentalization-based approaches (Bateman & Fonagy), may help mothers gain further awareness into how their parenting has been affected by their diagnosis, and how their behaviors may influence their children.

Parenting challenges

Mothers discussed several parenting challenges, which we coded into 4 categories. Of the 23 women we interviewed, 2 mothers in different groups discussed challenges with custody. Additionally, when asked about barriers preventing their participation in treatment that focused on their parenting, several mothers raised concerns about legal consequences or Child Protective Services (CPS) reports unfolding after discussing some of their poorer parenting moments in treatment. Unfortunately, it is unknown how many mothers with BPD are involved with CPS. However, it is possible that mothers with BPD interface with CPS at a higher rate than mothers without BPD. One study found a significantly higher rate of BPD symptoms in a sample of CPS-involved mothers than non CPS-involved mothers (Perepletchikova et al., 2012). Clinicians working with mothers who have BPD should be prepared to interface with the CPS offices in their state.

Not surprisingly, the most prominent thread throughout all of the interviews was mothers’ worry, guilt, and uncertainty around the parenting role. These findings are consistent with those reported by Newman and colleagues (2007), who found that mothers with BPD expressed less satisfaction and confidence and reported more distress about their parenting as compared to mothers without BPD. In the current study, mothers with BPD often showed intense guilt about trying to parent with so many mental health issues. Additionally, many mothers articulated uncertainty as to whether some of their emotions and frustrations with parenting were common to everyone or were a result of having mental health issues. Our team also noted that some mothers did not seem to realize that it is normative for all mothers to have some guilt, worry, or uncertainty associated with the parenting role. It is also possible that the guilt, worry, and uncertainty mothers with BPD possess are another manifestation of their symptom presentation. There are a variety of ways to address some of these concerns. One such way is to normalize the mothers’ emotions and thoughts surrounding parenting concerns through validation and empathy, which is an integral set of therapist skills emphasized within DBT but also recognized as common factors across empirically supported treatments for BPD (Bliss & McCardle, 2013; Weinberg et al., 2011). However, as articulated by several authors, therapists do not want to risk falling on either side of the dialectic, as in “pathologizing normative behavior” or “normalizing pathological behavior” (Bliss & McCardle, 2013; Kernberg, 1984; Miller et al., 2007).

Although mothers expressed uncertainty and often voiced not knowing how to handle parenting situations, some mothers expressed a clear desire to avoid certain parenting behaviors. These mothers shared memories from their upbringing of parenting styles they did not want to replicate. These stories concerned both how they were raised and how their own parents managed psychiatric issues within the family. Some mothers declared they would parent in the complete opposite manner than did their own parents. For instance, if a mother felt that no emotional expression was allowed in her childhood, she may allow her own child to express any emotion, anywhere. Although there is no empirical evidence to support that this series of transgenerational parenting patterns occur within BPD families, it is possible that women with BPD might be more likely to “swing” to another parenting extreme. Indeed, empirical evidence suggests that women with BPD tend to vacillate between hostile and helpless states of mind with respect to attachment figures (Lyons-Ruth, Melnick, Patrick, & Hobson, 2007), and this pattern is likely to play out in their parenting behaviors. In regards to integrating parenting and DBT and considering potential mechanisms of change, dialectical thinking could be developed by encouraging mothers with BPD to look for a synthesis between these extremes, which has been noted in a previous clinical case study (Ben-Porath, 2010). Non-DBT approaches for BPD (e.g., TFP, Kernberg, 1984; mentalization-based therapy, Bateman & Fonagy, 2004) may be similarly effective in helping mothers to integrate disparate mental states and representations with respect to themselves in relation to their children, perhaps building a more nuanced and coherent behavioral repertoire as parents. Additionally, future research efforts should employ methodologies such as parent-child observational tasks or interview mothers about their ‘meta-emotion’ philosophies (Gottman, Katz, & Hooven, 1996; Katz, Maliken, & Stettler, 2012) to uncover how mothers’ think about the emotion development of their children.

Last, mothers shared many examples of parenting being a burden, with some mothers reporting that they perceived their children as ‘clingy,’ which is possible given that insecure attachment styles are reported in research on infant and children with BPD mothers (Hobson et al., 2005; Macfie, 2009a). This is consistent with the data collected showing that mothers’ in our study experienced parenting as stressful. This finding reminds us that the effects of BPD and parenting are potentially bi-directional. BPD may negatively affect parenting, however, the role of parenting may exacerbate symptoms of BPD, especially for those mothers’ who find the role stressful. Future empirical research should examine whether reducing parenting stress predicts a reduction in BPD symptoms.

Parent-Child DBT

Mothers were asked a series of questions regarding whether and how they would prefer to have their current DBT-informed treatment modified to incorporate parenting issues. The following summarizes clinical considerations that should be taken into account for treatment development purposes. Mothers offered many ideas regarding the structure or format for how best to incorporate parenting treatment into their current DBT treatment. The various suggestions primarily had to do with whether mothers would bring their own children in with them, during treatment and whether children would attend treatment with the mother. Furthermore, if children were engaged in treatment, some mothers’ expressed a desire to have their children learn DBT skills. Although DBT is currently being adapted for pre-adolescent children (Perepletchikova et al., 2011), it is unclear whether teaching DBT skills to offspring of mothers with BPD is warranted as many children will not exhibit behavior and/or emotional problems. An empirical question remains as to whether teaching DBT skills to young children could reduce the incidence of future emotional and behavioral problems. Conversely, it is also possible that children’s behavior problems would decrease if their mother’s symptoms improved, as has been observed in the maternal depression literature (Gunlicks & Weissman, 2008).

Another barrier mentioned in regards to bringing children to mothers’ psychotherapy was womens’ occasional concern about other womens’ behavior. The mothers we interviewed stated that they would be concerned about bringing their child to treatment if they felt their child could witness highly dysregulated behaviors by other women. Additionally, given that parents have different values and some may choose to disclose more or less information regarding their diagnosis to their children, mothers also expressed concern about what their children may learn from other children in the group. These important issues would need to be addressed between the clinician and the patients prior to starting group treatment.

Finally, although not discussed with the participants, parenting issues would most likely be considered ‘Quality of Life’ behaviors, which are the third target in the DBT treatment hierarchy (Linehan, 1993a). Although an individual therapist would focus on life threatening and therapy interfering behaviors, this does not preclude a skills trainer from helping mothers use DBT skills within their parenting. DBT therapists treating mothers with BPD should help women find ways to generalize these skills to all domains of their life, including the maternal role. The overall findings from this qualitative study revealed that mothers with BPD in psychiatric treatment would benefit from some assessment and intervention integration around the role of parenting.

Strengths and Limitations

This is the first qualitative study to examine how mothers with BPD symptoms who are currently in psychiatric treatment experience the role of parenting and to ask women about their treatment preferences. The women who participated shared many personal stories and some had clear ideas for how their current DBT-informed treatment could be improved or adapted to address parenting explicitly. Despite the study strengths, there were several notable limitations to this work. This study was limited by not having conducted clinical diagnostic assessments on the women to verify that all of them met diagnostic criteria for a BPD diagnosis. However, given that these women were recruited from hospital programs that primarily treat women with BPD, and on average, women scored above the clinical cut-off for BPD symptoms on a self-report screening measure, we are confident that these women either had a BPD diagnosis or had highly elevated BPD symptoms. There were additional limitations associated with several aspects of heterogeneity of the sample. First, mothers who participated in this study had been in DBT-informed treatment for varying lengths. It is possible that this affected women’s thoughts and opinions about their treatment preferences, depending on what impairments in their life had been addressed in treatment to date. Secondly, the age of mother’s children ranged from parenting of infants to the parent-child relationship with adult offspring. How maternal BPD symptoms affect children will undoubtedly be related to the child’s age. A final study limitation is that we did not have a comparison group of either non-disordered mothers or mothers who had a different diagnosis. Therefore, the results from this qualitative study should be seen as reflective, although perhaps not unique to mothers with BPD.

Summary

This qualitative study identified mothers’ with BPD treatment concerns regarding parenting issues. Overall, women expressed a desire to explicitly target the parent-child relationship in their current psychiatric treatment. Future treatment development efforts should attempt to incorporate these preferences and aim to merge effective treatments for BPD with known effective parenting principles. These efforts should be undertaken because there is currently a treatment gap in that the role of being a parent is not explicitly addressed when treating women who have a serious and pervasive mental health issues, such as BPD. Although several efficacious treatments have been developed for BPD that appear to share some common putative mechanisms of change (Weinberg et al., 2011), none of these approaches explicitly address parenting issues in those with BPD. It is our belief that integrating parenting into current treatments may help mitigate an area of stress and burden for these women and could potentially even reduce the likelihood that the offspring go on to develop psychiatric symptoms themselves.

Table 1.

Summary of Domains, Categories, and Frequencies of the Experiences of Mothers’ with Borderline Personality Disorder who are Engaged in Psychiatric Treatment

Domains/Categories	Frequency
Concern with telling child
Developmental concerns	Typical
Not knowing how to communicate	Typical
Impact of BPD on children
Shared traits	Typical
Role changes/reversals	Typical
Associated life stressors	Variant
Emotion dysregulation	Typical
Parenting challenges
Custody issues	Variant
Guilt/worry/uncertainty	General
Reflections on own childhood	Variant
Burden of parenting	Typical

Note. General refers to all groups (9), Typical refers to 5–8 groups, and Variant means 2–4 groups.

Table 2.

Summary of Domains, Core Ideas, Categories, and Frequencies of Treatment Development Ideas for Parent-Child DBT

Domains/Categories	Frequency
Positive aspects of PC-DBT
Commonality	Typical
Children could benefit from learning skills	Typical
Increase Confidence as a parent	Typical
Potential barriers and clinical considerations with PC-DBT
Fear of judgment of own parenting	Typical
Fear of exposing children to other BPD Parents	Variant
Group structure	General
Timing	Typical
Logistics	Typical

Note. General refers to all groups (9), Typical refers to 5–8 groups, and Variant means 2–4 groups.

Appendix A: Interview

A. Understanding of the extent to which their children are aware of their mental illness

• 1.Are your children aware that you have been diagnosed with borderline personality disorder?
• 2.How have your children been affected by borderline personality disorder?

B. Connections between borderline personality disorder and any difficulties children may have

• 3.What are the connections between your experience of borderline personality disorder symptoms and any difficulties your children may have?

C. Parenting challenges that may be specific to living with borderline personality disorder

• 4.What challenges of parenting may be specific to their experience of borderline personality disorder.
• 5.How do you communicate with your children about borderline personality disorder?

D. Views on new Parent-Child Dialectical Behavior Therapy Program

Note: Before discussing views on the acceptability of Parent-Child Dialectical Behavior Therapy (PC-DBT), participants will be provided with a brief description of standard DBT and shown a plan of the new treatment and the topics that will be covered.

• 6.
  We plan to adapt standard DBT for adults for mothers who have a diagnosis of BPD and have children who are under the age of 5. How would you feel about taking part in such a treatment program and why?

  • 6a.
    Which parts of the treatment program may be easy/difficult?
  • 6b.
    How would you feel about focusing on parenting and the parent-child relationship in treatment?
  • 6c.
    How would you feel about having your child involved in treatment with you?

D1. Views on current DBT program

• 7.
  How would you change or improve the current DBT program to focus on parenting issues and your parent-child relationship?

  • 7a.
    Think about what would be actually discussed and skills taught. Is there anything you would change to better focus on parenting and the parent-child relationship?

D2. Views on the design and delivery of Parent-Child DBT

• 8.
  How would you design the Parent-Child DBT program to give you the most confidence in your parenting and to improve your relationship with your child?

  • 8a.
    Think about the timing. When would be the best time to have the group?
  • 8b.
    Think about where the treatment is conducted. Where would be the best place to have individual therapy and group?
  • 8c.
    What would get in the way of your attendance and participation in Parent-Child DBT?
  • 8d.
    What would increase your likelihood of attending and participating in Parent-Child DBT?