- Research article
- Open Access
Affect in response to stressors and coping strategies: an ecological momentary assessment study of borderline personality disorder
Borderline Personality Disorder and Emotion Dysregulationvolume 4, Article number: 8 (2017)
Affect instability is a core symptom of borderline personality disorder (BPD). Ecological momentary assessment allows for an understanding of real-time changes in affect in response to various daily stressors. The purpose of this study was to explore changes in affect in response to specific stressors and coping strategies in subjects with BPD utilizing ecological momentary assessment (EMA) methodology.
Subjects (n = 50) with BPD were asked to complete real-time assessments about stressors experienced, affect felt, and coping strategies employed six times per day for a 1-week period. Mixed effect regression models were used to measure the effect of stressors and coping strategies on affect change.
While most stressors led to experiencing more negative affect, only being in a disagreement was independently associated with increased negative affect. Among coping strategies, only doing something good for oneself independently reduced negative affect, controlling for all other coping strategies used.
These findings provide valuable insights into affective instability in BPD and can help inform treatment with individuals with the disorder.
Borderline personality disorder (BPD) is characterized by a pervasive pattern of instability in interpersonal relationships, self-image and emotion regulation that begin by early adulthood . Difficulty in affect regulation is often identified as a core deficit in BPD [2–6]. Individuals with BPD display a wide arrange of negative emotions, from rage to sadness to shame. They often experience rapid mood transitions from one intense emotion to another, thereby experiencing numerous negative affects throughout the course of a given day . Furthermore, once in an intense emotional state, individuals with BPD often find it difficult to recover . Thus, evidence-based interventions for the treatment of BPD place a strong emphasis on acquiring effective coping strategies to counter and manage rapid emotional changes .
Much of what we know about affective instability and emotion dysregulation in BPD is based on ratings of recollections and self-report trait questionnaires. Such retrospective ratings are considered to be unreliable and sensitive to reporter bias [9, 10]. Recall bias among individuals with BPD appears to be negatively skewed; patients with BPD tend to overestimate the intensity of negative affect and underestimate the intensity of positive affect during retrospective mood ratings, while healthy volunteers have the opposite recall bias . Ecological momentary assessment (EMA) provides a solution to this problem by allowing for real-time in vivo data collection of thoughts, feelings and behaviors . EMA data collection is characterized by four key features: 1) data are only collected on very recent state to minimize recall bias; 2) repetitive measurements are taken to quantify how a given phenomenon changes over time; 3) the phenomena is measured at predefined intervals or situations; and 4) measurement takes place in the individual’s natural environment . The use of EMA methodology has helped to develop a better understanding of BPD symptoms, including affective instability [11, 13–18]; dissociation ; interpersonal difficulties [20–22]; and suicidality [23–25].
Several EMA studies have shown that individuals with BPD exhibit greater variability and intensity of affect compared to both psychological healthy individuals [15, 17, 26] and other clinical populations [13, 27]. Specifically, individuals with BPD reported both a higher frequency and intensity of emotions than healthy controls during a 24-h psychophysiological ambulatory monitoring task, a form of EMA, where they were prompted every 10 to 20 min while awake to answer how they felt at the moment and the intensity of the feeling . Individuals with BPD reported more negative affect and less positive affect than their healthy control counterparts, and also reported greater intensity of negative affect, but not positive affect, during the study. Compared to healthy controls, those with BPD were more likely to rapidly fluctuate from a positive to a negative mood state . The affective instability found in individuals with BPD also distinguished them from other psychiatric populations . When examined together, these findings lend strong support to the unique emotional dysregulation and affective lability struggles faced by those with borderline personality disorder.
These studies, however, did not capture what environmental factors contributed to the affective instability reported in their samples. In a study that used event-contingent sampling and paper-and-pencil diaries to understand variability in mood in response to interpersonal interactions in a sample of patients with BPD, individuals with BPD reported greater variability in positive affect following social interactions that lasted at least 5 min compared to healthy controls . However, they reported comparable levels of variability in negative affect. While details on the participants’ interpersonal behavior were captured, information about the quality of the interaction and the participants’ feelings about the interaction (e.g., feeling rejected) were not. Participants with BPD also reported greater emotional reactivity in response to daily stress compared to patients with psychosis and healthy volunteers . However, participants were only able to report on the single most stressful event that occurred during each time period, leaving the possibility that affective variability and dysregulation caused by other stressors during that time period were overlooked. Therefore, further research is needed to better understand the interplay between the environment and affect instability in BPD.
The present study sought to evaluate the emotional reaction to specified stressors by regular sampling throughout the day. How subjects with BPD respond to specific stressors, with what emotions and how they manage their response and emotions may help guide therapeutic interventions.
Fifty participants who met DSM-IV-TR criteria for Borderline Personality Disorder were enrolled in this study. All participants provided written consent after receiving a detailed description of study procedures. The study protocol was approved by the Institutional Review Board at the New York State Psychiatric Institute. Eligibility criteria for the study included a diagnosis of borderline personality disorder and a history of either suicidal behavior or non-suicidal self-injury (at least one episode in the past 6 months and another within the past 2 years). Participants with bipolar I disorder, psychotic disorder, mental retardation, or an acute condition that required priority treatment, such as anorexia nervosa or severe substance dependence, were excluded. No comorbid Axis II diagnoses were excluded. The mean age of participants was 30.6 years (SD = 11, range: 18 to 62). As indicated in Table 1, the sample was predominantly female, Caucasian, and had a current diagnosis of Major Depressive Disorder. Eighty percent of the sample had a history of at least one suicide attempt, with an average of 2.2 suicide attempts (SD = 2.0) among attempters.
All participants completed a battery of assessments upon study entry with highly-trained Masters level psychologists who participated in regular reliability and consensus diagnostic conferences. Axis I and II psychopathology was diagnosed using the Structured Clinical Interviews for DSM-IV [28, 29]. Inter-rater reliability was high among assessors (ICC = 0.864) across Axis I and II disorders. The Beck Depression Inventory , a self-report measure of depressive symptoms was given to measure severity of depression at time of assessment.
Ecological momentary assessment
Participants were provided with a personal digital assistant (PDA) and completed a brief orientation session with a member of the research team. Participants practiced using the PDA with the assistant present until they were able to use the device on their own. They were instructed to carry the PDA with them at all times during a 1-week period and answer a series of questions when prompted. In order to best capture each participants’ day-to-day experiences, participants were asked to provide a 12-h period during which they expected to be awake and engaged in routine daily activities. During orientation, the PDA was customized to prompt each participant during their selected 12-h window. These prompts occurred on a random interval basis during a 12-h period. The daily 12-h window was divided into six two-hour blocks and one moment was randomly selected from each of these blocks for a total of six prompts over the course of each 12-h time period. Prompts were given randomly in order to avoid having participants anticipate when they would be asked to respond and to also ensure that a participant’s fixed schedule (e.g., work commitments) did not interfere with data collection on a regular basis.
At each prompt, participants were asked how strongly they experienced a series of affects since the last prompt using a 5-point Likert scale. The list of affects was derived from the Positive and Negative Affect Scales (PANAS)  and composite scores for positive and negative affect were calculated. In addition, participants were asked how overwhelmed they were by their feelings and to what degree their feelings felt out of control, using the same 5-point Likert scale. Participants were also asked whether any of the following stressors took place since the last prompt: 1) disagreement; 2) rejection; 3) compliment; 4) interpersonal disappointment; 5) neglect; 6) loss; 7) good news; 8) bad news; and 9) painful reminder from the past. Participants were able to indicate whether none, one, or several of these stressors took place during the assessment period. Finally, participants were asked if they used any of the following common coping strategies to manage the negative thoughts, feelings, or experiences they had since the last prompt: 1) keeping busy; 2) socializing; 2) positive thinking; 3) doing something good for self; 4) calming self; 5) finding perspective; and 6) sitting with feelings until they pass. The EMA prompts are detailed in Table 2. Similar to the questions about stressors, participants were able to indicate whether none, one, or several of the coping skills were employed. Participants also indicated whether they believed the coping strategies utilized were effective.
We analyzed the psychometric properties of the negative and positive affect scales from the EMA data following the recommendations of Shrout and Lane . Our analysis is informed by the fact that the prompts occurred randomly, making observation time essentially nested within the subject, and also we have unequal number of observations per subject. Analysis of the internal consistency of the two affect scales over time was performed using random effect models to separate the proportion of variability explained by between subject variation, item by subject interaction, and time nested within subject. Due to the large number of items in each scale, for item-wise analyses only summaries are presented.
For our main analyses, we assessed the effect of stressors and coping strategies on the change in positive and negative affect using mixed effect regression models. Change in positive affect at a given time t was measured by the difference between the positive affect score at time t to the positive affect score at the previous time period, (t-1), as long as both observations occurred on the same day. All analyses were performed using proc glimmix in the SAS™ software version 9.3 (Copyright@2002–2011, SAS Institute Inc., Cary, NC, USA), and the functions lme  and lmer in the statistical language R, version 2.12.1 . Mixed effect regression model was fit with positive affect change as outcome, the time-varying covariate (s) as predictor, subject-specific random intercepts, and correlations between observations within the same subject modeled using an AR1 structure. An identical method was employed for negative affect. Time-varying predictors were stressors during the respective time interval; frequencies of coping skills employed during the respective time period, and self-rated effectiveness of coping skills. In secondary or sensitivity analyses, stressors were recoded in the form of change scores, namely, a new stressor in any epoch compared to the previous epoch was coded as “+1”, no change in stressor as “0”, and a stressor disappearing was coded as “−1”. A second set of sensitivity analyses adjusted for the effect of the length of the time lag between prompts as a fixed effect, and then using continuous time correlation structure for the residuals. For the emotion dysregulation outcomes (feeling overwhelmed, feeling out of control), the measure was on a Likert scale and was treated as an ordinal variable. Mixed effect proportional odds logistic regression was fit with subject-specific random intercepts for these two outcomes. The predictors were event indicators (yes/no) and frequencies of coping skills applied. Age was examined as a covariate and was not found to be significant. Significance levels were not adjusted for multiple testing.
There were 1448 EMA records across subjects during the 7-day assessment period. On average, there were 29 records per subject, approximately 4 per day. The time interval between prompts answered within the same day had a median value of 2.1 h, with Inter Quartile Range of 1.3–3.9 h.
The average correlation between items in the Negative Affect Scale was 0.34 (range: 0.17–0.67). The average correlation of items with the total negative affect score was 0.56 (range: 0.45–0.68). The estimate of between-person reliability of the negative affect scale items, averaged over time in the EMA context, was 0.9995, while the reliability of within-subject change in negative affect in time was 0.8474. Item-wise, the proportion of variance explained by the differences between subjects as opposed to within subject change ranged from 31% (for “Upset” and “Irritable”) to 62% (for “Scared” and “Afraid”). For the Positive Affect Scale, the, average pair wise correlation, calculated from mixed effect models, between the 10 items was 0.39 (range: 0.24–0.64); the mean correlation with the scale total was 0.60 (range: 0.57–0.65). The estimate of between-person reliability of the positive affect scale items, averaged over time in the EMA context, was 0.9991, while the reliability of within-subject change in positive affect in time was 0.8733. The inter-correlation of the two subscales, adjusted for intra-subject correlations, to be r = −0.21, (t = −9.14, df = 1327, p < 0.0001).
Stressors as predictors of change in affect
We calculated change scores for negative and positive affect for observations within the same day, yielding a total of 979 measures for each. At each time point, participants reported an average of 2 stressors, with a range from 0 to 8, out of 9 possible stressors. Table 3 indicates the frequency of stressors reported, with “being reminded of something painful from the past” as the most commonly reported stressor.
In the mixed effect regression model with multiple predictors, having a disagreement (B = 3.43; t = 5.92; df = 920; p < 0.0001), was found to be associated with increased negative affect, controlling for all other stressors. Further, receiving a compliment was associated on a trend level with decreased negative affect (B = −0.92; t = −1.89; df = 920; p = .0596). Single-predictor models for six of the seven negative stressor categories, disagreement, rejection, interpersonal disappointment, neglect, bad news, and a painful reminder from the past, were associated with significant increase in negative affect, while neither of the two positive stressors, receiving a compliment and receiving good news, was significantly associated with a change in negative affect (Table 3).
With regard to increasing positive affect, one of the two positive stressors, receiving good news, was found to be significant in both the multiple predictor (Good News: B = 1.95; t = 3.43, df = 920, p = 0.0006) and individual predictor models (B = 2.09; t = 3.92; df = 928; p < 0.0001;). No negative stressors were found to lead to significant decrease in positive affect in the individual or multiple predictor models.
Sensitivity analyses using a coding that separated new stressors from no change supported all the significant results above, additionally, some of the stressors gained significance; specifically, new incidents of interpersonal disappointment (B = 1.63; t = 3.53; df = 920; p = 0.0004), receiving bad news (B = 2.23; t = 4.63; df = 920; p < 0.0001), being reminded of something painful from the past (B = 1.71; t = 4.21; df = 920; p < 0.0001) were found to be associated with increased negative affect, and receiving a new compliment increased positive affect (B = 1.77; t = 4.02, df = 920, p < 0.0001 after controlling for all other events. A second set of sensitivity analyses tested the effect of the length of the time interval between consecutive prompts on the change in affect. First, we adjusted our joint model for testing the effect of life events on (negative) affect change by the time lag. The effect of time lag was not significant (b = −1.40, SE = 0.85, z = −1.65, p = 0.10); and neither of the life events’ effects changed substantially. Second, we let the intra-subject correlation in the mixed effect model vary based on the time lag; again, the results stayed substantially the same, preserving significance where detected in the primary analysis.
Emotion dysregulation following a stressor
Participants were asked how overwhelmed they felt by their emotions, on a Likert scale from 1 to 5. In the mixed effect proportional odds logistic regression model with multiple predictors, four stressors were found to independently be associated with increased feelings of being overwhelmed: disagreement (OR = 2.70, t = 6.39, df = 1302, p < 0.0001); neglect (OR = 1.75, t = 3.81, df = 1302, p = .0001); bad news (OR = 1.92, t = 4.02, df = 1302, p < .0001); and painful reminder (OR = 2.08, t = 5.53, df = 1302, p < .0001). The two positive stressors were associated with decreased feelings of being overwhelmed in this model. Single-predictor models for each of the nine stressor categories indicated that all seven of the negative stressors were associated with an increase in feeling overwhelmed, while each of the positive stressors was associated with a decrease in feeling overwhelmed (Table 4).
Participants were also asked to what extent they felt a loss of control over their emotions. In the mixed effect proportional odds logistic regression model with multiple predictors, five stressors were found to be independent risk factors for feeling a loss of emotional control: disagreement (OR = 2.63, t = 6.10, df = 1302, p < 0.0001); interpersonal disappointment (OR = 1.43, t = 2.11, df = 1302, p = 0.0348); neglect (OR = 3.06, t = 3.16, df = 1302, p = .0016); bad news (OR = 1.44, t = 2.18, df = 1302, p = 0.0298); and painful reminder (OR = 1.79, t = 4.21, df = 1302, p < .0001). Again, separate single-predictor models for each of the nine stressor categories indicated that all seven of the negative stressors were associated with feeling a loss of emotional control (Table 5), while each of the positive stressors was associated with a decrease in feeling this way.
Coping strategies as predictors of decreased negative affect
Participants used on average 3.8 coping strategies per interval (S.D. = 2.4, range: 0 to 7). “Kept myself busy” was the most frequently used coping strategy (79%), while “calmed myself down” was the least commonly used (49%). The number of coping strategies applied was not associated with the length of time since the previous prompt (b = −0.33, t = −1.61, df = 914, p = 0.1078). Mixed effect regression models were used to test the effectiveness of each coping strategy on decreasing negative affect. In the multivariate model controlling for all of the other coping strategies, doing something good for oneself (B = −0.58, t = −2.54, df = 908, p = 0.0114) was the only significant independent predictors of decreased negative affect. In the single predictor models, two of the seven strategies were effective in decreasing negative affect: positive thinking (B = −0.34; t = −2.15; df = 915; p = 0.0321); and doing something good for self (B = −0.47; t = −2.97; df = 915; p = 0.0031). Sensitivity analyses showed that, when new coping strategies were separated out, positive thinking (B = −0.69, t = 0.13, df = 892, p = 0.0024) also had independent protective effect, while calming self (B = 0.80, t = 2.96, df = 892, p = 0.0032) and finding perspective (B = 0.89, t = 3.64, df = 892, p = 0.0003) both significantly increased negative affect.
Coping strategies as predictors of increased positive affect
Mixed effect regression models were also used to test the effectiveness of each coping strategy on increasing positive affect. Several coping skills independently increased positive affect in the multiple predictor model, including, positive thinking, doing something good for oneself, and finding perspective (see Table 5). Four of the seven strategies were effective in increasing positive affect in the single predictor models: keeping busy, socializing, positive thinking, and doing something good for oneself.
While EMA methodology has been used previously to better understand the symptomatology of BPD, to our knowledge, this is the first study to utilize EMA methodology to explore in real-time both stressors that contribute to changes in affect and the coping strategies employed to manage the emotional response to these stressors in a sample of individuals with BPD. Participants faced a substantial number of daily stressors, the most common of which was a painful reminder of something from the past, which may be partially attributable to the substantial lifetime history of physical and sexual abuse in this sample. Other frequent stressors included feeling neglected and facing an interpersonal disappointment. The frequency of daily stressors suggests that individuals with BPD, with or without concurrent depression, face substantial pain and distress in their everyday lives.
We found increases in negative affect in response to most of the negative stressors; however, only disagreement was independently associated with increased negative affect, after controlling for all other stressors, although six of the seven negative stressor categories were associated with increased negative affect in single predictor models. Stressors associated with feeling overwhelmed were having a disagreement, feeling neglected, receiving bad news, and being reminded of something painful from the past. The aforementioned stressors all remained independent risk factors for feeling out of control, along with being disappointed by someone. Because the stressors occurred during the same time period as feeling overwhelmed or out of control, it cannot be determined whether the feeling or the stressor came first. Receiving good news was found to significantly predict increased positive affect in both individual and joint predictor models, serving as a reliability check for the survey responses. The sheer number of stressors associated with increased negative affect, decreased positive affect and emotion dysregulation speaks again to how stressful daily life can be for an individual with BPD. Several of the stressors associated with negative affect and emotion dysregulation are interpersonal in nature. This is not surprising given that interpersonal sensitivity is considered to trigger both the impulsive behaviors and the emotion dysregulation that are at the core of BPD [6, 35–38]. Consistent with our findings, social rejection and negative evaluation have been previously found to contribute to affective instability in individuals with BPD [37, 38].
Individuals with BPD employed a range of coping skills to manage their emotions. The most commonly employed strategies were keeping busy, finding perspective, and positive thinking. Positive thinking and doing something good for self both independently improved positive affect. Doing something good for self also independently reduced negative affect. These findings gives credence to the emotion regulation strategies encouraged by Dialectical Behavior Therapy and other similar therapies, which place an emphasis on self-soothing and self-care in response to managing difficult emotions . The coping strategies of calming self, finding perspective, and sitting with feelings while they passed, while frequently used, were not helpful in this sample,. These strategies, particularly sitting with feelings until they pass, are similar to the mindfulness practice that is often prescribed for the treatment of BPD. However, it must be noted that these were individuals who were not in treatment and were not taught the skills in therapy; proper mindfulness training may lead to a more positive outcome with the use of this coping strategy, as indicated in other studies. Thus, our findings suggest that clinicians should be cautious when recommending this skill to their patients with BPD without adequate training. While it is not apparent why the coping strategies were more successful at increasing positive affect than decreasing negative fact, it is encouraging to find that these participants with untreated BPD were able to reduce their distress and improve their mood when faced with the numerous stressors they experienced on a daily basis.
This study is limited in several ways. The use of a 12-h window for data collection allows for the possibility that certain stressors, such as those occurring late at night, may not have been captured accurately. Our design of random prompts approximately every two hours does not allow for the detection of those affective changes that occur and disappear on a substantially shorter interval. The compliance rate for the overall sample during the weeklong assessment period was 69% (1,448 completed records from 2,100 prompts issued). While it is not known why participants answered the prompts at certain times and not others, it is possible that participants did not answer either when in crisis or when in a better mood, leading to the possibility of skewed results. Further, this study is limited to a BPD-only sample with no comparison control group. Thus, findings cannot be generalized to other clinical or psychiatrically healthy populations. Without exploring changes in affect in different clinical samples including psychiatrically healthy individuals, it is difficult to determine whether the response to stressors and coping strategies is unique to patients with BPD. This is particularly salient given that affective instability is not unique to BPD [39–42].
This study provides a preliminary understanding of how stressors contribute to changes in affect in individuals with BPD, as well as how strategies used by these individuals help them cope with affective changes. Further research is required to determine whether the changes in affect captured by EMA in this study in response to stressors or the coping strategies employed are unique to individuals with BPD, both in terms of frequency of affect change and intensity of affect experienced. In addition, future studies should continue to utilize real-time data to better understand the experiences of individuals with BPD in their actual environments, and potentially use this knowledge to tailor individualized treatments targeting their particular stressors and response to coping strategies.
Borderline personality disorder
Ecological momentary assessment
Personal digitial assistant
APA. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington: DSM-5; 2013.
Carpenter RW, Trull TJ. Components of emotion dysregulation in borderline personality disorder: a review. Current psychiatry reports. 2013;15(1):1–8.
Corrigan F, Davidson A, Heard H. The role of dysregulated amygdalic emotion in borderline personality disorder. Med Hypotheses. 2000;54(4):574–9.
Sanislow CA, Grilo CM, Morey LC, Bender DS, Skodol AE, Gunderson JG, McGlashan TH. Confirmatory Factor Analysis of DSM-IV Criteria for Borderline Personality Disorder: Findings From the Collaborative Longitudinal Personality Disorders Study. Am J Psychiatr. 2002;159(2):284–90. doi:10.1176/appi.ajp.159.2.284.
Skodol AE, Gunderson JG, Pfohl B, Widiger TA, Livesley WJ, Siever LJ. The borderline diagnosis I: psychopathology, comorbidity, and personaltity structure. Biol Psychiatry. 2002;51(12):936–50.
Stanley B, Siever LJ. The interpersonal dimension of borderline personality disorder: toward a neuropeptide model. Am J Psychiatry. 2009;167(1):24–39.
Koenigsberg HW, Harvey PD, Mitropoulou V, Schmeidler J, New AS, Goodman M, et al. Characterizing affective instability in borderline personality disorder. Am J Psychiatr. 2002;159(5):784–8.
Linehan MM. Cognitive-behavioral treatment of borderline personality disorder. New York: The Guilford Press; 1993.
Shiffman S, Stone AA, Hufford MR. Ecological momentary assessment. Annu Rev Clin Psychol. 2008;4:1–32.
Stone AA, Schwartz JE, Neale JM, Shiffman S, Marco CA, Hickcox M, Cruise LJ. A comparison of coping assessed by ecological momentary assessment and retrospective recall. J Pers Soc Psychol. 1998;74(6):1670.
Ebner-Priemer UW, Kuo J, Welch SS, Thielgen T, Witte S, Bohus M, Linehan MM. A valence-dependent group-specific recall bias of retrospective self-reports: A study of borderline personality disorder in everyday life. J Nerv Ment Dis. 2006;194(10):774–9.
Stone AA, Shiffman S. Ecological momentary assessment (EMA) in behavorial medicine. Ann Behav Med. 1994;16:199–202.
Trull TJ, Solhan MB, Tragesser SL, Jahng S, Wood PK, Piasecki TM, Watson D. Affective instability: Measuring a core feature of borderline personality disorder with ecological momentary assessment. J Abnorm Psychol. 2008;117(3):647.
Jahng S, Solhan M, Tomko R, Wood P, Piasecki T, Trull T. Affect and alcohol use: an ecological momentary assessment study of outpatients with borderline personality disorder. J Abnorm Psychol. 2011;120:572–84.
Ebner-Priemer UW, Kuo J, Kleindienst N, Welch SS, Reisch T, Reinhard I, Bohus M. State affective instability in borderline personality disorder assessed by ambulatory monitoring. Psychol Med. 2007;37(07):961–70.
Ebner-Priemer U, Kuo J, Schlotz W, Kleindienst N, Rosenthal M, Detterer L, et al. Distress and affective dysregulation in patients with borderline personality disorder: A psychophysiological ambulatory monitoring study. J Nerv Ment Dis. 2008;196:314–20.
Reisch T, Ebner‐Priemer U, Tschacher W, Bohus M, Linehan M. Sequences of emotions in patients with borderline personality disorder. Acta Psychiatr Scand. 2008;118(1):42–8.
Sadikaj G, Russell J, Moskowitz D, Paris J. Affect dysregulation in individuals with borderline personality disorder: persistence and interpersonal triggers. J Pers Assess. 2010;92:490–500.
Stiglmayr C, Ebner-Priemer U, Bretz J, Behm R, Mohse M, Lammers C, et al. Dissociative symptoms are positively related to stress in borderline personality disorder. Acta Psychiatr Scand. 2008;117:139–47.
Russell JJ, Moskowitz D, Zuroff DC, Sookman D, Paris J. Stability and variability of affective experience and interpersonal behavior in borderline personality disorder. J Abnorm Psychol. 2007;116(3):578–88.
Stepp S, Pilkonis P, Yaggi K, Morse J, Feske U. Interpersonal and emotional experiences of social interactions in borderline personality disorder. J Nerv Ment Dis. 2009;197:484–91.
Berenson K, Downey G, Rafaeli E, Coifman K, Paquin N. The rejection-rage contingency in borderline personality disorder. J Abnorm Psychol. 2011;120:681–90.
Links P, Eynan R, Heisel M, Barr A, Korzekwa M, McMain S, et al. Affective instability and suicidal ideation and behavior in patients with borderline personality disorder. J Personal Disord. 2007;21:72–86.
Links P, Eynan R, Heisel M, Nisen- Baum R. Elements of affective instability associated with suicidal behaviour in patients with borderline personality disorder. Can J Psychiatr. 2008;53:112–6.
Nisenbaum R, Links P, Eynan R, Heisel M. Variability and predictors of negative mood intensity in patients with borderline personality disorder and recurrent suicidal behavior: Multilevel analyses applied to experience sampling methodology. J Abnorm Psychol. 2010;119:433–9.
Ebner-Priemer UW, Welch SS, Grossman P, Reisch T, Linehan MM, Bohus M. Psychophysiological ambulatory assessment of affective dysregulation in borderline personality disorder. Psychiatry Res. 2007;150(3):265–75.
Glaser J-P, Os JV, Mengelers R, Myin-Germeys I. A momentary assessment study of the reputed emotional phenotype associated with borderline personality disorder. Psychol Med. 2008;38(09):1231–9.
Santangelo P, Bohus M, Ebner-Priemer UW. Ecological momentary assessment in borderline personality disorder: a review of recent findings and methodological challenges. J Personal Disord. 2014;28(4):555–76.
First MB, Spitzer RL, Gibbon M, Williams JB. Structured clinical interview for DSM-IV-TR axis I disorders, research version, patient edition: SCID-I/P. 2002.
Beck AT, Ward C, Mendelson M. Beck depression inventory (BDI). Arch Gen Psychiatry. 1961;4(6):561–71.
Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol. 1988;54(6):1063.
Shrout P, Lane SP. Psychometrics. In: Mehl MR, Conner TS, editors. Handbook of research methods for studying daily life. 2012. p. 302–20.
Pinheiro J, B. D., DebRoyS, Sarkard D, R Core Team. (2015). nlme: Linear and Nonlinear Mixed Effects Models_. R package version 3.1–119, from < URL:http://CRAN.R-project.org/package=nlme >.
The R Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing. Retrieved from http://www.R-project.org/.
Gunderson JG. Disturbed relationships as a phenotype for borderline personality disorder. American J of Psychiatry. 2007;164(11):1637–40.
Gunderson JG, Lyons-Ruth K. BPD’s interpersonal hypersensitivity phenotype: a gene-environment-developmental model. J Personal Disord. 2008;22(1):22–41.
Gratz KL, Rosenthal MZ, Tull MT, Lejuez CW, Gunderson JG. An experimental investigation of emotional reactivity and delayed emotional recovery in borderline personality disorder: The role of shame. Compr Psychiatry. 2010;51(3):275–85.
Gratz KL, Dixon-Gordon KL, Breetz A, Tull M. A laboratory-based examination of responses to social rejection in borderline personality disorder: the mediating role of emotion dysregulation. J Personal Disord. 2013;27(2):157–71.
Henry C, Mitropoulou V, New AS, Koenigsberg HW, Silverman J, Siever LJ. Affective instability and impulsivity in borderline personality and bipolar II disorders: similarities and differences. J Psychiatr Res. 2001;35(6):307–12.
Henry C, Van den Bulke D, Bellivier F, Roy I, Swendsen J, M’Baïlara K, Leboyer M. Affective lability and affect intensity as core dimensions of bipolar disorders during euthymic period. Psychiatry Res. 2008;159:1–6.
Marwaha S, Gordon-Smith K, Broome M, Briley PM, Perry A, Forty L, Jones L. Affective instability, childhood trauma and major affective disorders. J Affect Disord. 2016;190:764–71.
Kashdan TB, Uswatte G, Steger MF, Julian T. Fragile self-esteem and affective instability in posttraumatic stress disorder. Behav Res Ther. 2006;44(11):1609–19.
Funded by 1P50MH090964-01A1 and 5R01MH061017-10.
Availability of data and materials
The datasets during and/or analyzed during the current study available from the corresponding author on reasonable request.
HG analyzed the data and SC, HG, BS and JM interpreted the data. EB collected the EMA data and assisted in data analysis. SC, HG and BS were major contributors in writing the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
This study was approved by the New York State Psychiatric Institute (NYSPI) Institutional Review Board. All participants provided informed written consent prior to beginning study procedures.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.