The Course of Marriage/Sustained Cohabitation and Parenthood among Borderline Patients Followed Prospectively for 16 Years

Abstract

The purpose of this study was to determine the rate of marriage/sustained cohabitation and parenthood reported by recovered and non-recovered borderline patients, the age first undertaken, and the stability of these relationships. Borderline patients were interviewed about these topics during their index admission and eight times over 16 years of prospective follow-up. Recovered borderline patients were significantly more likely than non-recovered borderline patients to have married/lived with an intimate partner and to have become a parent. In addition, they first married/cohabited and became a parent at a significantly older age. They were also significantly less likely to have been divorced or ended a cohabiting relationship. In addition, they were significantly less likely to have given up or lost custody of a child. Taken together, the results of this study suggest that stable functioning as a spouse/partner and as a parent are strongly associated with recovery status for borderline patients.

Marriage and having children have typically been seen as markers of a good adult adaptation. They join a steady work record and the ability to support oneself as markers of the autonomy and competence that many in society expect from adults and that many adults have as personal goals. In addition, parents often worry that their children will be alone and/or seen as different if they do not marry and have children.

These milestones apply to those with borderline personality disorder (BPD) and those with other psychiatric disorders as much as they apply to psychiatrically healthy adults. Despite their centrality to the aspirations and concerns of many of those with BPD and their family members, this topic has not been the subject of much research. Two epidemiological studies have assessed these outcomes in community samples of men and women in the US meeting criteria for BPD. Swartz et al. (1990) using data from the Duke site of the Epidemiological Catchment Area project found that 24% of the 1.8% (N=24) of subjects meeting rough criteria for BPD were married and living with their spouse and 11% were separated or divorced. Tomko et al. (2013) reanalyzed Grant's National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) data (2008) and found 2.7% met stringent DSM-IV criteria for BPD. Of these 1,030 subjects, 45% were married/cohabitating and 31% were separated, divorced, or widowed.

Seventeen small-scale short-term studies of the prospective course of BPD have been published (Zanarini, Frankenburg, Hennen, & Silk, 2003). Of these 17 studies, only five provided data on rates of marriage/cohabitation (Grinker, Werble, & Drye, 1968; Modestin & Villiger, 1989; Links, Mitton, & Steiner, 1990; Sandell et al., 1993; Antikainen, Hintikka, Lehtonen, Koponen, & Arstila, 1995) and only one provided the rate of having children (Modestin & Villiger, 1989). Rates of marriage ranged from a low of 22% to a high of 54%, with a median of 37%. Only Modestin and Villiger reported the rate of being a parent in their sample, which was followed for 4 ½ years—33%.

Four large-scale long-term follow-back studies of the course of BPD have also been published (Plakun, Burkhardt, & Muller, 1985; McGlashan, 1986; Paris, Brown, & Nowlis, 1987; Stone, 1990). Only two of these four studies, which assessed outcomes a mean of 14–16 years post index admission, reported rates of marriage and parenthood (McGlashan, 1986; Stone, 1990). McGlashan, in the Chestnut Lodge study, found that 70% had ever been married and 48% had children. Stone, who studied patients who were hospitalized at the New York State Psychiatric Institute, found the following lifetime rates: marriage (45%) and parenthood (23%). Paris followed his Montreal sample again at a mean of 27 years and found that 83% had been married or cohabitating with someone at some point in their life and 59% had one or more children (Paris & Zweig-Frank, 2001).

More recently, there have been two NIMH-funded prospective studies of the longitudinal course of borderline personality disorder (Zanarini, Frankenburg, Hennen, Reich, & Silk, 2005; Skodol et al., 2005). At their six-year follow-up, Zanarini et al. found that 32% of the borderline patients in the McLean Study of Adult Development (MSAD) were married or living with someone and 27% were parents (Zanarini, Frankenburg, Reich, Hennen, & Silk, 2005). At their final assessment period (10-year follow-up), Gunderson et al. (2011) found that 41% of borderline subjects in the Collaborative Longitudinal Personality Disorders Study (CLPS) were either married or living with someone.

The current study is the first prospective study to assess the rates of marriage/sustained cohabitation and parenthood in a rigorously diagnosed group of borderline patients over 16 years of prospective follow-up. Rates of ending these relationships through divorce or breaking up, and losing or giving up custody of one's children were also assessed. In addition, the age at which these events first occurred was ascertained.

Methods

The current study is part of the McLean Study of Adult Development (MSAD), a multifaceted longitudinal study of the course of borderline personality disorder. The methodology of this study, which was reviewed and approved by the McLean Hospital Institutional Review Board, has been described in detail elsewhere (Zanarini, Frankenburg, Hennen, & Silk, 2003). Briefly, all subjects were initially inpatients at McLean Hospital in Belmont, Massachusetts. Each patient was screened to determine that he or she: 1) was between the ages of 18–35; 2) had a known or estimated IQ of 71 or higher; 3) had no history or current symptomatology of schizophrenia, schizoaffective disorder, bipolar I disorder, or an organic condition that could cause serious psychiatric symptoms, such as multiple sclerosis or lupus erythematosus; and 4) was fluent in English.

After the study procedures were explained, written informed consent was obtained. Each patient then met with a masters-level interviewer blind to the patient's clinical diagnoses for a thorough psychosocial functioning/treatment history and diagnostic assessment. Four semistructured interviews were administered: 1) the Background Information Schedule (BIS) (which assesses psychosocial functioning and treatment history) (Zanarini, 1992); 2) the Structured Clinical Interview for DSM-III-R Axis I Disorders (SCID I) (Spitzer, Williams, Gibbon, & First, 1992); 3) the Revised Diagnostic Interview for Borderlines (DIB-R) (Zanarini, Gunderson, Frankenburg, & Chauncey, 1989); and 4) the Diagnostic Interview for DSM-III-R Personality Disorders (DIPD-R) (Zanarini, Frankenburg, Chauncey, & Gunderson, 1987). The inter-rater and test-retest reliability of all four of these measures have been found to be good-excellent (Zanarini, Frankenburg, Reich et al., 2005; Zanarini & Frankenburg, 2001; Zanarini, Frankenburg, & Vujanovic, 2002).

At each of eight follow-up assessments, separated by 24 months, axis I and II psychopathology were reassessed via interview methods similar to the baseline procedures by staff members blind to baseline diagnoses. After informed consent was obtained, our diagnostic interview battery was re-administered. The Revised Borderline Follow-up Interview—the follow-up analog to the Background Information Schedule—was also administered at each of these eight waves of follow-up (Zanarini, Sickel, Yong, & Glazer, 1994). The follow-up inter-rater reliability (within one generation of follow-up raters) and follow-up longitudinal reliability (from one generation of raters to the next) of these four measures have also been found to be good-excellent (Zanarini, Frankenburg, Reich et al., 2005; Zanarini & Frankenburg, 2001; Zanarini et al., 2002).

Data pertaining to marriage/sustained intimate relationships, their dissolution, becoming a parent, and loss or relinquishment of custody were obtained primarily from the BIS and BFI-R. A brief interview specifically designed to assess these aspects of psychosocial functioning over the first 16 years of follow-up was also administered to serve as a validity check.

Statistical Analyses

We compared borderline patients who had recovered over the 16 years of prospective follow-up to those who did not on eight variables related to intimate relationships and parenthood. We aggregated data across time in order to present overall percentages, Ns, means, and standard deviations. Chi-squared tests were used to compare recovered and non-recovered borderline patients on categorical variables and Student's t-tests were used in comparisons of continuous variables. Given the relatively large number of comparisons, we applied the Bonferroni correction for multiple comparisons. This resulted in the following adjusted alpha level of significance: 0.05/8= 0.006.

Results

Two hundred and ninety patients met both DIB-R and DSM-III-R criteria for borderline personality disorder. In terms of baseline demographic data, 80.3% (N=233) of the subjects were female and 87.2% (N=253) were white. The average age of the borderline subjects was 26.9 years (SD=5.8), their mean socioeconomic status was 3.4 (SD=1.5) (where 1=highest and 5=lowest), and their mean GAF score was 38.9 (SD=7.5) (indicating major impairment in several areas, such as work or school, family relations, judgment, thinking, or mood).

In terms of continuing participation, 87.5% (N=231/264) of surviving borderline patients were re-interviewed at all eight follow-up waves. Of the 26 who died, 13 committed suicide and 13 died of other causes.

One hundred and fifty borderline patients met previously defined criteria for recovery and 140 did not (Zanarini, Frankenburg, Reich, & Fitzmaurice, 2010). We had selected a Global Assessment of Functioning (GAF) score of 61 or higher as our measure of two-year recovery because it offers a reasonable description of a good overall outcome (i.e., some mild symptoms or some difficulty in social, occupational, or school functioning, but generally functioning pretty well, has some meaningful interpersonal relationships). We operationalized this score to enhance its reliability and meaning. More specifically, to be given this score or higher, a subject typically had to have achieved a symptomatic remission, have at least one emotionally sustaining relationship with a close friend or life partner/spouse, and be able to work or go to school consistently, competently, and on a full-time basis (which included being a house-person) during the two-year interval.

All told, 59.7% of borderline patients (N=173) had either married and/or cohabited with someone for a period of five years or more. In terms of being a parent, 40.3% had had a child or raised one (N=117).

Table 1 compares borderline patients who have achieved recovery and those who had not on eight variables related to enduring intimate relationships and parenthood. Recovered borderline patients were significantly more likely to have married or had a relationship with a partner with whom they lived for five years or more than non-recovered borderline patients (79% vs. 39%). They were also significantly older when first entering this type of relationship (29 vs. 25). They were significantly more likely to have had or raised a child (49% vs. 31%). They were also significantly older when first doing so (30 vs. 23). However, they were significantly less likely to dissolve a long-standing intimate relationship (42% vs. 75%) and significantly less likely to lose or relinquish custody of their child or children (7% vs. 51%). Age at first dissolution (early 30s) and loss of custody (late 20s) was about the same for those in both the recovered and non-recovered groups.

Table 1.

Aspects of Marriage/Cohabitation and Parenting among Recovered and Non-recovered Borderline Patients Followed Prospectively for 16 Years

	Ever Recovered BPD (%/N or Mean/SD)	Never Recovered BPD (%/N or Mean/SD)	χ2/t-test	p-value
Ever Married/Lived with Intimate Partner for Sustained Period of Time	78.7% (N=118)	39.3% (N=55)	46.7	<0.001
Age at First Marriage/Sustained Cohabitation	29.0 (SD=6.6)	25.2 (SD=7.0)	3.40	<0.001
Ever Divorced/Ended Sustained Cohabitation	42.4% (N=50)	74.6% (N=41)	15.6	<0.001
Age at Divorce/End of Sustained Cohabitation	31.5 (SD=6.3)	32.8 (SD=7.1)	0.94	0.352
Ever Had or Adopted Child	49.3% (N=74)	30.7% (N=43)	10.4	0.001
Age at First Child's Birth or Adoption	29.6 (SD=7.6)	22.6 (SD=5.6)	5.26	<0.001
Ever Lost/Relinquished Custody of Child/Children	6.8% (N=5)	51.2% (N=22)	30.2	<0.001
Age Lost or Relinquished Child/Children	29.0 (SD=11.4)	28.6 (SD=7.4)	0.10	0.920

We also assessed the role of gender in these analyses pertaining to marriage/sustained partnership and parenthood. More specifically, we compared women with borderline personality disorder who had and had not recovered. We also compared men with borderline personality disorder who had and had not recovered; because only 20% of the subjects were male, some of the comparisons should be interpreted with caution due to sparseness of data.

In terms of women, recovered women were significantly more likely than non-recovered women to have married or lived with an intimate partner for five years or more (79.2% vs. 42.6%, χ²=33.0, p<0.001) and to have had or adopted children (52.8% vs. 31.5%, χ²=10.7, p=0.001). They were also significantly less likely to have divorced or ended a sustained cohabiting relationship (42.4% vs. 80.4%, χ²=18.3, p<0.001) and to have lost or relinquished custody of their child or children (4.0% vs. 16.7%, χ²=10.4, p=0.001). In addition, they were older when they first married or lived with someone for a sustained period of time (although not significantly older at the Bonferroni-corrected level) (28.6 vs. 25.4, t-test=2.57, p=0.01) and they were significantly older when they first had or adopted a child (29.0 vs. 22.2, t-test=4.52, p<0.001).

In terms of men, recovered men were significantly more likely than non-recovered men to have married or lived with an intimate partner for five years or more (76.0% vs. 28.1%, χ²=12.9, p<0.001). They were also significantly older when they first married or lived with someone for a sustained period of time (30.8 vs. 24.2, t-test=3.02, p=0.0056). However, the rates of divorce or ending a sustained cohabiting relationship (42.1% vs. 44.4%) and the rates of having had or adopted a child (32.0% vs. 28.1%) were very similar for both the recovered and non-recovered men. Of those recovered men who had or adopted a child, they were significantly older than non-recovered men when first becoming a parent (35.0 vs. 24.2, t-test=4.79, p=0.002). There was also a trend for non-recovered men to be more likely than recovered men to lose or relinquish custody of their child or children (12.5% vs. 0.0%, χ²=3.4, p=0.067).

Discussion

Two main findings have emerged from this study. The first is that recovered borderline patients were significantly more likely to marry or live with a partner for a sustained period of time and to become a parent than borderline patients who had not experienced a concurrent two-year symptomatic remission and good social and vocational functioning. They were also significantly older when first embarking on such a relationship and becoming a parent. These results are not surprising as those who are less ill may well have more emotional energy to invest in their relationships and more opportunities to meet appropriate potential partners due to their better psychosocial functioning.

The second is that recovered borderline patients were significantly more likely than non-recovered borderline patients to have stability in these areas of their personal lives. More specifically, they were significantly less likely to divorce or break up with an intimate partner and significantly less likely to lose or relinquish custody of their children. While they were about half as likely to end an intimate relationship, they were seven times less likely to lose or relinquish custody of their children. These results too are not surprising as many borderline patients report having stormy close relationships (Zanarini, Frankenburg, Gunderson, & Chauncey, 1990). This storminess may arise from within, be the result of assortative mating, or a combination of the two. The older age of recovered borderline patients at the time they first married or entered a long-term cohabitation relationship and their older age when first becoming a parent suggests that they gave themselves more time to address their problems and to mature in a number of areas of functioning before undertaking these responsibilities than non-recovered borderline patients.

These two main findings apply as well to recovered and non-recovered women as to the whole sample of recovered and non-recovered patients with borderline personality disorder. This is not surprising as the high percentage of women in our sample may have driven our overall results. However, the pattern reported by men was somewhat different. Like recovered women, recovered men were significantly more likely to have entered into some type of live-in romantic relationship that was sustained. They were also significantly older when first entering this type of intimate relationship and when first having children. In addition, non-recovered men were substantially more likely to have lost or given up custody of their child or children; however, this difference was not statistically discernable due to a lack of statistical power. Unlike women, both groups of men with borderline personality disorder were about as likely to have been divorced or ended a sustained cohabiting relationship. They were also equally likely to have had or adopted a child.

In comparison to the whole sample or women with borderline personality disorder, the rates of divorce/ending a sustained live-in romantic relationship are similar to those reported by those in the ever recovered group and the rates of becoming a parent are similar to those in the never recovered group. Taken together, these results suggest that recovery status appears to have a weaker association with these two aspects of men's personal lives than it does for women. The reasons for this are unclear. It may be that men with borderline personality disorder are more content in their intimate relationships or have more supportive partners than women with BPD. It may also be that men with BPD regardless of their recovery status are less likely to become a parent than women with borderline personality disorder.

Some clinicians believe that the intimate relationships of most borderline patients end in divorce or a chaotic breakup. They also believe that most borderline patients cannot be good parents. Our results suggest that recovered borderline patients who, as noted above, are older when marrying or moving in with a romantic partner and when first becoming a parent can achieve stability in these areas.

It is also important to note that even the borderline patients who had trouble parenting stably due to their continuing illness, in most cases voluntarily relinquished custody of their children to a relative rather than having a government agency remove their child or children due to neglect or abuse. In most cases, they visited regularly and in many cases, their young adult children have returned to their home to live— suggesting a somewhat transient difficulty with parenting due to the severity of their illness and resultant psychosocial impairment rather than an inherent lifelong deficit in these areas.

It should be noted that the rates of marriage/cohabitation and parenthood we found for borderline patients in general are consistent with the rates found by McGlashan (1986). It should also be noted that they are somewhat higher than those found by Stone (1990) and substantially lower than those found by Paris in his 27 year follow-up (Paris & Zweig-Frank, 2001). The reasons for these differences are not clear but may be related to age at study onset, the length of follow-up, or some admixture of the two. More specifically, the borderline patients in the current study were, on average, about the same age as those in the Chestnut Lodge study and somewhat older than those in the New York State Psychiatric Institute study at study entry. In addition, the current study assessed these outcomes after 16 years of prospective follow-up, while the second follow-up of Paris occurred a mean of almost 30 years after study entry.

This study has two main limitations. One limitation of this study is that all of the patients were seriously ill inpatients at the start of the study. Another limitation is that about 90% of those in both patient groups were in individual therapy and taking psychotropic medications at baseline and about 70% were participating in each of these outpatient modalities during each follow-up period (Hörz, Zanarini, Frankenburg, Reich, & Fitzmaurice, 2010). Thus, it is difficult to know if these results would generalize to a less disturbed group of patients or people meeting criteria for borderline personality disorder who were not in treatment, which was typically non-intensive outpatient treatment as usual in the community.

Taken together, the results of this study suggest that stable functioning as a spouse/partner and as a parent are strongly associated with recovery status for borderline patients.