Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: Personal Disord. 2016 Aug 15;9(1):73–80. doi: 10.1037/per0000211

Borderline Personality Pathology and Physical Health: The Role of Employment

Patrick J Cruitt 1, Michael J Boudreaux 1, Joshua J Jackson 1, Thomas F Oltmanns 1
PMCID: PMC5311027  NIHMSID: NIHMS805114  PMID: 27657166

Abstract

Borderline personality disorder (BPD) is associated with negative physical health outcomes. Clinical case studies suggest that employment status may buffer against the negative effects of BPD on physical health. The goal of the current study was to examine the interaction between BPD features and employment status in predicting subjective perceptions of physical health. We hypothesized that employment status would moderate the relationship between BPD features and physical health, such that individuals who are employed would exhibit a weaker negative relationship between BPD features and self- and informant ratings of physical health. We investigated this question using data from a community sample of 1,630 middle-aged to older adults participating in the St. Louis Personality and Aging Network, an ongoing study of personality, health, and aging. Results indicated that employment status and BPD features were significant predictors of both self- and informant ratings of physical health. Confirming our hypothesis, the interaction term contributed to a significant increase in the proportion of explained variance, suggesting that employment is associated with a weaker negative relationship between BPD features and physical health. These findings highlight the importance of examining occupational functioning in the long-term course of BPD and offer avenues for further research into moderators of the relationship between BPD features and physical health.

Keywords: personality disorder, borderline personality disorder, physical health, employment, long-term outcomes

The deleterious impact of personality pathology on physical health outcomes is well established in the literature, and symptoms associated with borderline personality disorder (BPD) appear to be particularly harmful (Dixon-Gordon, Whalen, Layden & Chapman, 2015). One study found that a diagnosis of BPD was associated with increased rates of cardiovascular disease, arthritis, gastrointestinal disease and several other medical conditions in a nationally representative sample (El-Gabalawy, Katz & Sareen, 2010). In addition, BPD features are associated with worse concurrent physical functioning, as well as increased healthcare utilization and medication usage at a six-month follow-up (Powers & Oltmanns, 2012).

Despite the harmful impact of BPD features on health-related outcomes, few studies have examined factors that might buffer against these effects. Clinical case studies of remitted BPD patients suggest that having a supportive relationship, religious faith, or fulfilling job may provide these individuals with a source of self-esteem and emotional support, potentially leading to an improvement in functioning (Paris, 2003). One promising implication of these case studies is that competent and consistent performance in an occupational setting may interact with BPD symptoms in a beneficial manner, such that individuals with BPD symptoms who are employed exhibit better outcomes than those who are unemployed.

Previous research on occupational functioning and BPD has focused almost exclusively on a somewhat different issue, i.e. the extent to which occupational functioning is impaired in individuals with BPD features. One study found that symptoms of BPD were associated with lower levels of education, as well as an increased likelihood of experiencing conflict at work, dismissal or demotion, and unemployment (Hengartner, Müller, Rodgers, Rossler, & Ajdacic-Gross, 2014). The McLean Study of Adult Development, a longitudinal study of the course of BPD, found that poor psychosocial functioning persists beyond the remission of symptoms in many patients with BPD (Zanarini, Frankenburg, Reich, & Fitzmaurice, 2010a; Zanarini, Frankenburg, Reich, & Fitzmaurice, 2012). Although 99% of the patients in their sample experienced a symptomatic remission of at least two years over the course of the study, only 60% achieved a recovery of psychosocial functioning (Zanarini et al., 2012). Recovery in the occupational domain seems to be especially difficult to attain. The primary reason that some patients did not meet the McLean Study's definition of recovery was due to poor occupational functioning rather than social functioning (Zanarini, Frankenburg, Reich, & Fitzmaurice, 2010b; Zanarini et al., 2012). These unrecovered BPD patients were more likely to suffer from chronic medical illnesses, have poor health-related behaviors, and over utilize health services when compared to participants who experienced a complete psychosocial recovery (Keuroghlian, Frankenburg, & Zanarini, 2013). As such, employment is a salient factor to consider when exploring the links between BPD features and poor physical health outcomes.

The relationship between employment, BPD features, and physical health outcomes may be especially important in the context of later life. Cross-sectional data indicate that certain BPD symptoms, particularly impulsivity and self-harm, are more prevalent in younger individuals with a BPD diagnosis (Blum et al., 2008). At the same time, worse health perceptions and increased health care utilization are more characteristic of older BPD patients (Blum et al., 2008). These findings suggest that the impact of BPD features on physical health could persist after the most prominent symptoms of the disorder may have faded. Older adults face a number of role transitions, such as retirement, that may exacerbate the expression of personality pathology. Chronic health problems also become more prevalent in older age, making it an ideal context for studying the relationship between personality pathology and physical health (Oltmanns & Balsis, 2011).

The current study used multiple sources and methods to assess BPD features. Previous research has relied primarily on self-report instruments or semi-structured diagnostic interviews for the measurement of BPD features, and therefore is limited by reliance on the self as the sole source of information. Individuals who exhibit BPD symptoms may be less likely or unable to report accurately on their experiences or on the effect that their behavior may have on others (Oltmanns & Turkheimer, 2009). Some evidence indicates that informant reports of personality pathology have predictive value above and beyond self-reports when examining physical health outcomes, such as coronary heart disease (Jackson, Connolly, Garrison, Leveille, & Connolly, 2015; Kneip, et al., 1993; Smith et al., 2007). Thus, obtaining information from multiple perspectives may enhance measurement validity for assessing personality pathology.

Measurement of personality pathology can also be improved by utilizing perspectives beyond the personality disorder (PD) types presented in the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5; American Psychiatric Association, 2013). Recent research has investigated links between the ten PD types and the five-factor model (FFM) of normal personality (Samuel & Widiger, 2008; Saulsman & Page, 2004). PD count scores have been developed for each of the ten PDs based on theoretical and empirical relationships with the lower-order facets of the Revised NEO Personality Inventory (Miller, Bagby, Pilkonis, Reynolds, & Lynam, 2005). By using both types of measurements, the current study is better able to assess subthreshold BPD symptoms, providing evidence regarding the impact of these features in a non-clinical population.

The purpose of this study was to examine associations among employment status, BPD symptoms, and perceptions of physical health. In particular, we hypothesized that employment status would moderate the relationship between symptoms of BPD and physical health, such that individuals who are employed would exhibit a weaker negative association between BPD features and physical health relative to those who are unemployed. We also hypothesized that this interaction effect would hold for both self- and informant-report of physical health.

Method

Participants and Procedure

The analyses reported here used data from the St. Louis Personality and Aging Network (SPAN), a longitudinal study examining personality, health, and aging in later life. Participants were a representative sample of adults living in the St. Louis area. At the time they entered the study, participants were between the ages of 55 and 64 (M = 59.5, SD = 2.7). For more details regarding sample characteristics and recruitment procedures, see Oltmanns, Rodrigues, Weinstein, and Gleason (2014). The current analyses used data from the baseline assessment. This assessment consisted of an informed consent process including a thorough explanation of the procedures, followed by a battery of interviews and questionnaires administered over a 3-hr session.

In total, 1,630 participants completed the baseline assessment. Of these, 55% were female (n = 889). Sixty-five percent were Caucasian (n = 1,060) and 32% were Black/African-America (n = 517). Due to small sample sizes in other categories, race was recoded into White/Caucasian and Black/African-American/Other (coded as 0 or 1, respectively) for the present analyses. Participants indicated their level of education by selecting one of nine discrete response options. These options were subsequently recoded to represent years of education, ranging from 6.5 to 20 (M = 15.3, SD = 2.6).

During the baseline assessment, we asked participants to nominate an informant who knew them well and could describe their personality. Overall, we recruited 1,467 informants. They were 68% female (n = 1,004) and were between the ages of 18 and 92 (M = 55, SD = 11.4). Roughly half of the informants were a spouse or romantic partner of the participant (48%, n = 702). The remaining were typically either other family members (28%, n = 404) or friends (22%, n = 328). On average, informants and participants had known each other for 32 years (SD = 15).

Measures

Structured Interview for DSM-IV Personality (SIDP-IV; Pfohl, Blum, & Zimmerman, 1997)

Trained interviewers administered the SIDP-IV to each participant in order to identify symptoms of personality pathology. The SIDP-IV is a semi-structured, diagnostic interview that consists of 80 items. The items correspond to each of the diagnostic criteria for the ten PD types. Ratings for each criterion are provided on a scale from 0 (not present) to 3 (strongly present). A criterion is considered endorsed when a rating of two or three is given; by this definition 27 participants met three or more BPD criteria. In order to obtain a continuous score of BPD features, the ratings that corresponded to each of the nine BPD criteria were summed. Scores ranged from 0 to 16 (out of a possible maximum of 27). The interviews were video-recorded, and independent judges rated a randomly selected subsample of 265 interviews. Inter-rater reliability was assessed using a one-way random, average measures intra-class correlation coefficient (Shrout & Fleiss, 1979). Overall, the reliability of the entire SIDP interview was .67, and the reliability for BPD was .77, indicating good to excellent agreement.

Multi-source Assessment of Personality Pathology (MAPP; Okada & Oltmanns, 2009; Oltmanns, Turkheimer, & Strauss, 1998)

The MAPP was developed to obtain both self- and informant reports of PD symptoms. The current version consists of 80 items, which are lay translations of the criteria for the 10 DSM-IV PDs. Participants and informants respond using a 5-point Likert scale from 0 (I am/he or she is never like this) to 4 (I am/he or she is always like this). The current analyses used only the BPD scale, which consists of 9 items. BPD scores on the self-report version of the MAPP ranged from 0 to 23 and scores on the informant report version ranged from 0 to 28 (out of a possible maximum of 36). Coefficient alpha for the self-report version of the MAPP BPD scale was .70, indicating adequate reliability. The informant report BPD scale also had adequate reliability, with a coefficient alpha of .79.

NEO-Personality Inventory-Revised (NEO PI-R; Costa & McCrae, 1992)

The NEO PI-R is a 240-item measure of the FFM. It assesses the five domains of neuroticism, extraversion, openness, agreeableness and conscientiousness, as well as six lower-order facets for each domain. We used both the self-report (Form S) and informant report (Form R) versions for the current study. In order to link the FFM to the PD literature, Miller et al. (2005) proposed a count technique that generates a continuous trait score for each of the 10 DSM-IV PDs using the facet scores of the NEO PI-R. Scores on the relevant facets are summed together, with facets being reversed-scored as needed. The NEO PD count score for BPD combines the facets of anxiety, angry hostility, depression, impulsiveness, vulnerability, openness to feelings, openness to actions, compliance (reverse scored) and deliberation (reverse scored). Items were rated on a 0 to 4 scale, and coefficient alphas for the facets ranged from .60 (compliance) to .82 (depression). The range for the BPD count score in the current sample was 63.63 to 84.13 for self-report, and 61 to 85.88 for informant report.

Rand-36 Health Status Inventory (HSI; Hays & Morales, 2001)

The HSI is a self-report questionnaire consisting of 36 items assessing subjective physical and mental health. The present analyses used the physical health composite of the HSI. This composite score is composed of four weighted subscales, including physical functioning, role limitations due to physical problems, pain, and general health perceptions. Coefficient alphas for these subscales ranged from .75 (pain) to .92 (physical functioning). Composite scores ranged from 29 to 72. Several studies have supported the reliability and validity of the RAND-36 HSI in community samples of adults (Moorer, Suurmeijer, Foets, & Molenaar, 2001; VanderZee, Sanderman, Heyink, & de Haes, 1996), and have shown that self-reported health is uniquely predictive of mortality, even when objective health factors are considered (Benyamini & Idler, 1999; Idler & Benyamini, 1997).

Informant-Report Version of the HSI

Because certain personality traits may lead participants to report more negative perceptions of their physical health, we adapted the items from the HSI that could be reasonably answered by an informant. This resulted in an informant report version of the HSI that included 10 out of the original 36 items. Four of these items assessed general and physical health status, and were combined to form an informant rated physical health status scale. These items were: “In general, would you say his/her health is...”, “Compared to 1 year ago, how would you rate his/her health in general now?”, “During the past 4 weeks, to what extent has his/her physical health interfered with his/her normal social activities with family, friends, neighbors, or groups?”, and “During the past 4 weeks, to what extent has his/her physical health interfered with his/her ability to work or engage in physical activity?” Each was rated on a 1 to 5 scale, with response labels differing between questions. Scores ranged from 4 to 20. Coefficient alpha for this 4-item scale was .77, indicating adequate reliability.

Employment

We obtained employment information through a demographic questionnaire. Participants indicated whether they were “currently employed or self-employed” using a dichotomous yes/no response. Sixty-two percent of the sample was employed, whereas 36% was unemployed. Data were missing for the remaining 2% of participants (n = 30), who were therefore not included in the analyses. We also asked whether the individual was “retired from a field other than the military.” Overall, 30.7% of the sample indicated that they were retired. Of these, 345 (21.2%) were not currently working, whereas 155 (9.5%) were currently employed in some capacity after retirement (this could include part-time work or work in a field other than the one from which the participant retired).

Data Analytic Plan

We derived standardized BPD component scores by submitting total scores on the SIDP and self- and informant reported MAPP and NEO PD count scores to a principal components analysis. In order to test the hypothesis that employment status moderates the relationship between BPD features and self- and informant rated health, we ran a series of hierarchical multiple regression analyses. In each analysis, employment status and BPD features were included as predictors of health status as well as their interaction, together with the covariates of race, gender, years of education, and household income. Prior to fitting the models, all variables were standardized.

Results

Table 1 presents the intercorrelations among the five BPD measures as well as employment and health status. The BPD measures were moderately to strongly correlated. All five measures of BPD features were negatively correlated with both self-reported physical health status and informant reported health status. They were uncorrelated or weakly negatively correlated with employment status. Employment status had a moderate positive correlation with self-reported physical health and informant reported health status.

Table 1.

Correlations between Demographics, Borderline Personality Disorder, and Physical Health Measures

M (SD) 1 2 3 4 5 6 7 8 9 10 11 12 13
Demographics
    1. Age 59.5 (2.7)
    2. Gendera 0.45 (0.5) −.02
    3. Raceb 0.34 (0.5) −.02 −.01
    4. Educationc 6.0 (2.1) −.04 .07 −.33
    5. Household Incomed 3.9 (2.2) −.05 .18 −.34 .48
    6. Employment 0.63 (0.5) −.18 −.01 −.16 .23 .25
Borderline PD
    7. SIDP-IV 1.1 (1.9) −.08 .02 .02 −.10 −.18 −.07
    8. MAPP – Self 3.9 (3.5) −.06 .17 −.02 −.09 −.07 −.09 .42
    9. MAPP – Informant 5.0 (4.8) −.09 .00 −.01 −.11 −.14 −.08 .35 .27
    10. NEO – Self 71.2 (2.7) −.08 −.06 −.08 −.09 −.16 −.06 .45 .54 .30
    11. NEO – Informant 71.9 (3.8) −.05 −.11 −.07 −.11 −.17 −.04 .36 .26 .68 .45
    12. BPD Factor --- −.10 .00 −.03 −.13 −.20 −.08 .70 .67 .72 .76 .77
Physical Health
    13. Self HSI 59.1 (10.0) .01 .06 −.25 .29 .34 .31 −.26 −.27 −.22 −.30 −.20 −.34
    14. Informant HSI 15.2 (3.1) −.02 .03 −.11 .19 .23 .21 −.20 −.17 −.32 −.21 −.32 −.33 .60
Open in a new tab

Note. N ranges from 1,183 to 1,600. Correlations greater than |.08| are significant at p < .001. SIDP-IV = Structured Interview for DSM-IV Personality; MAPP = Multi-source Assessment of Personality Pathology; NEO = NEO Personality Inventory-Revised; HSI = RAND-36 Health Status Inventory.

a

Gender coded as 1 = men, 0 = women

b

Race coded as 1 = African American, 0 = White

c

Median years of education is a 4-year college degree.

d

Median household income is between 40,000 and 59,999 a year.

Because the BPD measures shared overlapping variance, we submitted the intercorrelations among the measures to a principal components analysis.1 The scree plot indicated one general factor, which explained 52.80% of the variance. We therefore extracted a single component and computed component scores for all participants, which provided us with a weighted linear composite of the five BPD measures. Component loadings for these five measures ranged from .67 to .77.

We then performed a series of hierarchical regression analyses predicting self- and informant ratings of physical health. In the first step, we entered BPD features and employment status into the model, with age, gender, race, education, and household income2. The interaction between BPD and employment was entered in the second step. Cumulative R2 values and parameter estimates are reported in Tables 2 (for self-reported physical health) and 3 (for informant reported physical health).

Table 2.

Summary of Moderation Analysis Predicting Self-reported Physical Health Status

Predictor R 2 Unstandardized b SE(b) Standardized b SE(b) 95% CI t value
Step 1 .280
    Age 0.09 0.09 0.02 0.02 (−0.02; 0.07) 1.04
    Gender 0.28 0.47 0.01 0.02 (−0.03; 0.06) 0.60
    Race −3.31 0.54 −0.16 0.03 (−0.21; −0.11) −6.12***
    Education 0.39 0.13 0.08 0.03 (0.03; 0.13) 2.95**
    Household Income 0.77 0.13 0.17 0.03 (0.11; 0.22) 5.98***
    Employment 3.92 0.51 0.19 0.02 (0.14; 0.24) 7.70***
    BPD Features −3.65 0.36 −0.27 0.02 (−0.32; −0.23) -11.22***
Step 2 .285
    BPD × Employment 1.43 0.47 0.07 0.02 (0.02; 0.11) 3.01**
Open in a new tab

Note. Regression coefficient estimates based on final model.

N = 1327.

*p < .05.

**

p < .01.

***

p < .001.

Table 3.

Summary of Moderation Analysis Predicting Informant Reported Physical Health Status

Predictor R 2 Unstandardized b SE(b) Standardized b SE(b) 95% CI t value
Step 1 .162
    Age −0.01 0.03 −0.01 0.03 (−0.07; 0.04) −0.47
    Gender 0.04 0.17 0.01 0.03 (−0.05; 0.06) 0.25
    Race −0.30 0.19 −0.05 0.03 (−0.11; 0.01) −1.53
    Education 0.08 0.05 0.05 0.03 (−0.01; 0.12) 1.69
    Household Income 0.14 0.05 0.10 0.03 (0.03; 0.16) 2.98**
    Employment 0.83 0.18 0.13 0.03 (0.07; 0.19) 4.59***
    BPD Features −1.17 0.13 −0.28 0.03 (−0.34; −0.23) −10.05***
Step 2 .168
    BPD × Employment 0.48 0.17 0.07 0.03 (0.02; 0.13) 2.86**
Open in a new tab

Note. Regression coefficient estimates based on final model. N = 1142.

*p < .05.

**

p < .01.

***

p < .001.

As shown, BPD features and employment status were significant predictors of both self- and informant ratings of physical health. As predicted, the interaction term was significantly related to both outcomes, demonstrating that the magnitude of the BPD-physical health relationship was influenced by employment status (self-report: ΔR2 = .005, F [1, 1318] = 9.08, p = .003; informant report: ΔR2 = .006, F [1, 1133] = 8.15, p = .004). The nature of these interactions can be seen in Figures 1 and 2.3 As shown, the slopes were weaker for employed participants (standardized b = −.22, −.22) than unemployed participants (−.36, −.38) for both self and informant reports of health, respectively. This pattern indicates that the effects of BPD features on physical health were less pronounced for employed participants than for unemployed participants.

Figure 1.

Figure 1

Open in a new tab

BPD by Employment Interaction Predicting Self-Reported Health

Figure 2.

Figure 2

Open in a new tab

BPD by Employment Interaction Predicting Informant-Reported Health

Given the age range of the participants in the current study, it is not surprising that many had retired. Retirement is often voluntary, and therefore may not exhibit the same negative effect on the BPD-physical health relationship as unemployment for other reasons. As such, we ran two additional analyses excluding those who had retired from the “not currently employed” category, while continuing to include those who had retired but were still working in the “currently employed” category. As before, we estimated the same series of hierarchical regression models, except using these modified variables. We found that the interaction term (standardized b = .09 and .10, for self- and informant report of health, respectively) again contributed significant variance to the prediction of both self- (ΔR2 = .008, F [1, 1030] = 10.90, p < .001) and informant (ΔR2 = .008, F [1, 878] = 8.82, p = .003) ratings of physical health. Thus, employment status moderated the relationship between BPD features and physical health even when those participants who had retired were excluded.

Discussion

The results of this study support the hypothesis that employment status moderates the relationship between BPD symptoms and perceptions of physical health. Individuals who were employed exhibited a weaker negative relationship between BPD features and physical health relative to those who were unemployed. This relationship replicated across both self- and informant reports of physical health. By establishing a link between employment status, BPD symptoms and physical health outcomes, these findings offer preliminary support for investigating employment as a partial explanation for the better outcomes observed in some BPD patients (Paris, 2003).

This paper represents an encouraging initial effort to identify a factor that may help mitigate the negative impact associated with BPD features. Previous investigations, such as the McLean Study, have identified functional impairments associated with BPD symptoms in a variety of domains and clarified the long-term presence of psychosocial difficulties for individuals who exhibit features of BPD (Zanarini et al., 2010a; Zanarini et al., 2012). This is the first study to examine empirically the manner in which positive functioning in one particular domain is associated with a weaker relationship between BPD symptoms and their negative consequences. The fact that these results were obtained in the context of occupational functioning, which research has shown to be particularly impaired in individuals with clinically significant BPD symptoms, is notable (Zanarini et al., 2010b). Treatment interventions tend to focus on social functioning as the most salient domain involved in BPD symptoms. These analyses suggest that occupational functioning may also be an important domain to address during treatment. Although individuals need to acquire certain social skills to obtain and maintain employment, employment may also provide individuals with a consistent social context in which to develop these skills. Given that both individuals and society incur a large cost as the result of poor physical functioning associated with BPD (van Asselt, Dirksen, Arntz, & Severens, 2007), identifying factors that may offer opportunities to improve these outcomes is an important task for the field.

The analyses reported here do have a few limitations. First, the data are cross-sectional. As such, it is difficult to determine the directionality of the relationships examined in this study, as it is possible that people with the poorest health are unable to work or choose to retire sooner. Another potential limitation is the use of questionnaires to measure physical health. Although replicated with informant's reports of physical health, it will be important to look at biomarkers and other objective indicators of physical health in future analyses. Finally, the current study used a simple distinction between those participants who were “currently employed” and those who were “not currently employed.” More comprehensive assessments of occupational functioning may help uncover the mechanisms involved, especially when considering older adults navigating the transition into retirement. For example, measurements of occupational satisfaction and the importance of work to the individual may provide insight into the degree to which the beneficial impact of work on the BPD-health relationship is due to enhanced self-esteem or identity formation.

In addition to addressing the limitations of the current analyses, future research can expand on these results in a number of ways. Although analyses using data from SPAN consistently show that even subthreshold symptoms of BPD have a significant impact on physical health (Powers & Oltmanns, 2012), it is important to replicate the current analyses in clinical samples with a greater prevalence of individuals who meet a clinical cut off for BPD. Additionally, as the SPAN study continues collecting data, longitudinal analyses of the effect of employment status on the relationship between BPD features and physical health will become possible. Previous case studies suggest that engaging in a stable, rewarding career buffers against the negative effects of BPD features on health (Paris, 2003). Therefore, future studies should examine the degree to which individuals’ jobs are central to their identity and self-esteem, and whether these qualities strengthen the moderating effect of employment. Subsequent research could also investigate other factors that may moderate the BPD-physical health relationship. Potential additional factors identified by Paris (2003) include a stable romantic relationship and involvement with a religious faith. As these potential moderators are examined and mechanisms are identified, the therapeutic implications of these results will need to be explored.

The current paper connects to broader observations about the relationship between psychopathology and work in the context of agency (i.e., the capacity of individuals to act independently and to make their own free choices). Recovering a sense of agency appears to be a crucial task facing individuals suffering from many forms of psychopathology (Strauss & Davidson, 1997). Employment is a key domain in which agency may be developed and maintained, and it can provide a source of meaning and purpose in an individual's life. A case example from the SPAN study illustrates this possible function of employment. When interviewed at baseline, one of our participants who met full criteria for BPD appeared to be functioning remarkably well. She described her work as a teacher, and reported winning awards for working with students whom no one else would teach. This job clearly was a source of great pride for her, and her accomplishments undoubtedly helped her maintain an integration of her identity in later life. Of course, the same function may be accomplished through other means, such as volunteering, which previous research has shown may be one way in which conscientious individuals express the high-achieving aspects of their identity after retirement (Mike, Jackson, & Oltmanns, 2014). The conscious decision to work and the value one derives from one's work are crucial aspects of identity expression in Western society. As such, it is crucial to consider work when exploring the breakdown of identity formation associated with BPD symptoms, and the subsequent impact this has on an individual's life and physical health.

Acknowledgments

This work was supported by grants from the National Institute of Mental Health (RO1-MH077840-01) and National Institutes of Health (NIH 5 T32 AG000030-39).

The authors would like to thank Elizabeth Edershile, Janine Galione, Juliette McClendon-Iacovino, and Hannah King for their comments on the manuscript. They would also like to thank Merlyn Rodrigues for her help with data collection and management.

Footnotes

1

We also ran analyses with each BPD measure separately. There were no apparent systematic differences in clinician, self- or informant-report of BPD features, so we proceeded with the analyses using the component scores.

2

The results were very similar whether one controlled for these variables or not.

3

For the sake of clarity, and because the results were very similar, the figures represent the model without controlling for covariates.

References

  1. American Psychiatric Association . Diagnostic and statistical manual of mental disorders. 5th ed. Author; Washington, DC: 2013. [Google Scholar]
  2. Blum N, Franklin J, Hansel R, McCormick B, St. John D, Pfohl B, Black DW. Relationship of age to symptom severity, psychiatric comorbidity and health care utilization in persons with borderline personality disorder. Personality and Mental Health. 2008;2(1):25–34. doi: 10.1016/j.comppsych.2007.05.005. doi:10.1002/pmh.26. [DOI] [PubMed] [Google Scholar]
  3. Benyamini Y, Idler E. Community studies reporting association between self rated health and mortality, additional studies, 1995 to 1998. Research on Aging. 1999;21:392–401. [Google Scholar]
  4. Costa PT, McCrae RR. Revised NEO Personality Inventory (NEO PI-R) and the NEO Five-Factor Inventory (NEO-FFI) Professional Manual. Psychological Assessment Resources; Odessa, FL: 1992. [Google Scholar]
  5. Dixon-Gordon KL, Whalen DJ, Layden BK, Chapman AL. A systematic review of personality disorders and health outcomes. Canadian Psychology/Psychologie Canadienne. 2015;56(2):168–190. doi: 10.1037/cap0000024. doi:10.1037/cap0000024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. El-Gabalawy R, Katz LY, Sareen J. Comorbidity and associated severity of borderline personality disorder and physical health conditions in a nationally representative sample. Psychosomatic Medicine. 2010;72(7):641–647. doi: 10.1097/PSY.0b013e3181e10c7b. doi:10.1097/PSY.0b013e3181e10c7b. [DOI] [PubMed] [Google Scholar]
  7. Hays RD, Morales LS. The RAND-36 measure of health-related quality of life. Annals of Medicine. 2001;33(5):350–357. doi: 10.3109/07853890109002089. [DOI] [PubMed] [Google Scholar]
  8. Hengartner MP, Müller M, Rodgers S, Rössler W, Ajdacic-Gross V. Occupational functioning and work impairment in association with personality disorder trait-scores. Social Psychiatry and Psychiatric Epidemiology. 2014;49(2):327–335. doi: 10.1007/s00127-013-0739-2. doi:10.1007/s00127-013-0739-2. [DOI] [PubMed] [Google Scholar]
  9. Idler E, Benyamini Y. Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior. 1997;38:21–37. [PubMed] [Google Scholar]
  10. Jackson JJ, Connolly JJ, Garrison SM, Leveille MM, Connolly SL. Your friends know how long you will live: A 75-year study of peer-rated personality traits. Psychological Science. 2015;26(3):335–340. doi: 10.1177/0956797614561800. doi:10.1177/0956797614561800. [DOI] [PubMed] [Google Scholar]
  11. Keuroghlian AS, Frankenburg FR, Zanarini MC. The relationship of chronic medical illnesses, poor health-related lifestyle choices, and health care utilization to recovery status in borderline patients over a decade of prospective follow-up. Journal of Psychiatric Research. 2013;47(10):1499–1506. doi: 10.1016/j.jpsychires.2013.06.012. doi:10.1016/j.jpsychires.2013.06.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Kneip RC, Delamater AM, Ismond T, Milford C, Salvia L, Schwartz D. Self- and spouse ratings of anger and hostility as predictors of coronary heart disease. Health Psychology. 1993;12(4):301–307. doi: 10.1037//0278-6133.12.4.301. doi:10.1037/0278-6133.12.4.301. [DOI] [PubMed] [Google Scholar]
  13. Mike A, Jackson JJ, Oltmanns TF. The conscientious retiree: The relationship between conscientiousness, retirement, and volunteering. Journal of Research in Personality. 2014:5268–77. doi: 10.1016/j.jrp.2014.07.002. doi:10.1016/j.jrp.2014.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Miller JD, Bagby RM, Pilkonis PA, Reynolds SK, Lynam DR. A Simplified Technique for Scoring DSM-IV Personality Disorders with the Five-Factor Model. Assessment. 2005;12(4):404–415. doi: 10.1177/1073191105280987. doi:10.1177/1073191105280987. [DOI] [PubMed] [Google Scholar]
  15. Moorer P, Suurmeijer TM, Foets M, Molenaar IW. Psychometric properties of the RAND-36 among the chronic diseases (multiple sclerosis, rheumatic diseases and COPD) in the Netherlands. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care & Rehabilitation. 2001;10(7):637–645. doi: 10.1023/a:1013131617125. doi:10.1023/A:1013131617125. [DOI] [PubMed] [Google Scholar]
  16. Okada M, Oltmanns TF. Comparison of three self-report measures of personality pathology. Journal of Psychopathology and Behavioral Assessment. 2009;31(4):358–367. doi: 10.1007/s10862-009-9130-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Oltmanns TF, Balsis S. Personality disorders in later life: Questions about the measurement, course, and impact of disorders. Annual Review of Clinical Psychology. 2011;7:321–349. doi: 10.1146/annurev-clinpsy-090310-120435. doi:10.1146/annurev-clinpsy-090310-120435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Oltmanns TF, Rodrigues MM, Weinstein Y, Gleason MJ. Prevalence of personality disorders at midlife in a community sample: Disorders and symptoms reflected in interview, self, and informant reports. Journal of Psychopathology and Behavioral Assessment. 2014;36(2):177–188. doi: 10.1007/s10862-013-9389-7. doi:10.1007/s10862-013-9389-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Oltmanns TF, Turkheimer E. Person perception and personality pathology. Current Directions in Psychological Science. 2009;18(1):32–36. doi: 10.1111/j.1467-8721.2009.01601.x. doi:10.1111/j.1467-8721.2009.01601.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Oltmanns TF, Turkheimer E, Strauss ME. Peer assessment of personality traits and pathology in female college students. Assessment. 1998;5(1):53–65. doi: 10.1177/107319119800500108. [DOI] [PubMed] [Google Scholar]
  21. Paris J. Personality disorders over time: Precursors, course and outcome. American Psychiatric Publishing, Inc; Quebec, Canada: 2003. [DOI] [PubMed] [Google Scholar]
  22. Pfohl B, Blum N, Zimmerman M. Structured Interview for DSM-IV Personality. American Psychiatric Press; Washington, DC: 1997. [Google Scholar]
  23. Powers AD, Oltmanns TF. Personality disorders and physical health: A longitudinal examination of physical functioning, healthcare utilization, and health-related behaviors in middle-aged adults. Journal of Personality Disorders. 2012;26(4):524–538. doi: 10.1521/pedi.2012.26.4.524. doi:10.1521/pedi.2012.26.4.524. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Samuel DB, Widiger TA. A meta-analytic review of the relationships between the five-factor model and DSM-IV-TR personality disorders: A facet level analysis. Clinical Psychology Review. 2008;28(8):1326–1342. doi: 10.1016/j.cpr.2008.07.002. doi:10.1016/j.cpr.2008.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Saulsman LM, Page AC. The five-factor model and personality disorder empirical literature: A meta-analytic review. Clinical Psychology Review. 2004;23(8):1055–1085. doi: 10.1016/j.cpr.2002.09.001. doi:10.1016/j.cpr.2002.09.001. [DOI] [PubMed] [Google Scholar]
  26. Shrout PE, Fleiss JL. Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin. 1979;86(2):420–428. doi: 10.1037//0033-2909.86.2.420. doi:10.1037/0033-2909.86.2.420. [DOI] [PubMed] [Google Scholar]
  27. Smith TW, Uchino BN, Berg CA, Florsheim P, Pearce G, Hawkins M, Yoon H. Hostile personality traits and coronary artery calcification in middle-aged and older married couples: Different effects for self-reports versus spouse ratings. Psychosomatic Medicine. 2007;69(5):441–448. doi: 10.1097/PSY.0b013e3180600a65. doi:10.1097/PSY.0b013e3180600a65. [DOI] [PubMed] [Google Scholar]
  28. Strauss JS, Davidson L. Mental disorders, work, and choice. In: Bonnie RJ, Monahan J, editors. Mental disorder, work disability, and the law (105-130) University of Chicago Press; Chicago, IL: 1997. [Google Scholar]
  29. van Asselt ADI, Dirksen CD, Arntz A, Severens JL. The cost of borderline personality disorder: Societal cost of illness in BPD-patients. European Psychiatry. 2007;22:354–361. doi: 10.1016/j.eurpsy.2007.04.001. http://dx.doi.org/10.1016/j.eurpsy.2007.04.001. [DOI] [PubMed] [Google Scholar]
  30. VanderZee KI, Sanderman R, Heyink JW, de Haes H. Psychometric qualities of the RAND 36-Item Health Survey 1.0: A multidimensional measure of general health status. International Journal of Behavioral Medicine. 1996;3(2):104–122. doi: 10.1207/s15327558ijbm0302_2. [DOI] [PubMed] [Google Scholar]
  31. Zanarini MC, Frankenburg FR, Reich DB, Fitzmaurice G. Time to attainment of recovery from borderline personality disorder and stability of recovery: A 10-year prospective follow-up study. The American Journal of Psychiatry. 2010a;167(6):663–667. doi: 10.1176/appi.ajp.2009.09081130. doi:10.1176/appi.ajp.2009.09081130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zanarini MC, Frankenburg FR, Reich DB, Fitzmaurice G. The 10-year course of psychosocial functioning among patients with borderline personality disorder and axis II comparison subjects. Acta Psychiatrica Scandinavica. 2010b;122(2):103–109. doi: 10.1111/j.1600-0447.2010.01543.x. doi:10.1111/j.1600-0447.2010.01543.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Zanarini MC, Frankenburg FR, Reich DB, Fitzmaurice G. Attainment and stability of sustained symptomatic remission and recovery among patients with borderline personality disorder and axis II comparison subjects: A 16-year prospective follow-up study. The American Journal of Psychiatry. 2012;169(5):476–483. doi: 10.1176/appi.ajp.2011.11101550. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES