Andreja Pšeničny, Mitja Perat CAN BURNOUT RESULT FROM PERSONALITY DISORDERS? 1-23 human resources development institute gasilsk a 19 si-1000 ljubljana 2 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat izvirni znanstveni članek ::POVZETEK JE IZGOREVANJE LAHKO POSLEDICA OSEBNOSTNIH MOTENJ? Avtorji članka v svoji psihoterapevtski praksi opažajo, da diagnostična ocena osebnostne strukture pri večini pacientov s sindromom izgorelosti pokaže značilnosti mejne osebnostne organizacije srednjega nivoja, relativno bližje nevrotični strukturi, oziroma zmerno integrirane (mejne) osebnostne strukture. Zato smo ugotavljali, ali je izgorelost preko storilnostnega sam- ovrednotenja in deloholizma povezana z bordeline, narcistično in shizoidno mejno osebnostno motnjo. 3393 udeležencev je izpolnilo šest vprašalnikov (Vprašalnik sindroma adrenalne izgorelosti, Vprašalnik storilnostnega samovrednotenja, Vprašalnik deloholizma, ter presejalne teste za bordeline, shizoidno in narcistično mejno osebnostno motnjo). Pričakovali smo, da se bodo tisti, ki imajo več znakov bordeline, shizoidne in narcistične mejne osebnostne motnje, tudi bolj vrednotili po dosežkih in bodo bolj deloholični in bodo zato močneje izgorevali. Analiza variance za ponovljene meritve je potrdila naša empirična opažanja iz klinične psihoterapevtske prakse, da je med izgorelimi osebami večina takih, ki kažejo višjo stopnjo znakov vseh treh mejnih osebnostnih motenj, bordeline, shizoidne in narcistične. Potrdila je tudi, da sta storilnostno samovrednotenje in deloholizem tudi indikatorja prisotnosti teh motenj, saj imajo udeleženci, ki nimajo znakov izgorelosti, se pa vrednotijo po dosežkih oziroma so deloholični, višjo stopnjo znakov teh treh mejnih osebnostnih motenj. Diskriminatna analiza pa je potrdila tudi hipotezo, da so vse tri mejne osebnostne motnje ob kovariantah sam- ovrednotenja po dosežkih in deloholizmu ustrezni prediktorji izgorelosti, najmočnejši med njimi pa je bordeline osebnostna motnja. Ključne besede: izgorelost, storilnostno samovrednotenje, deloholizem, mejne osebnostne motnje ABSTRACT In their psychotherapy practice, the authors of this article observed that the diagnostic assessment of the personality structure in most patients with burnout syndrome shows features of medium-level borderline personality organization, relatively closer to a neurotic structure or moderately integrated borderline personality structure. Therefore, we examined whether burnout is associated with borderline, narcis- sistic, and schizoid borderline personality disorders through performance-based self-esteem and workaholism. A total of 3,393 respondents completed six question- naires (the Adrenal Burnout Syndrome Questionnaire, the Performance-Based 3 can burnout result from personality disorders? Self-Esteem Scale, the Work Addiction Risk Test, and screening tests for borderline, schizoid, and narcissistic personality disorders). We expected that those that had more symptoms of borderline, schizoid, and narcissistic borderline personality disorders would also tend to base their self-worth more on achievements, would be more workaholic, and would therefore show more burnout symptoms. A repeated measures analysis of variance confirmed our empirical observations from clinical psychotherapy practice that the majority of burned-out participants also show a higher level of symptoms of all three borderline personality disorders: borderline, schizoid, and narcissistic. It also confirmed that performance-based self-esteem and workaholism are indicators of the presence of personality disorders because respondents that have no burnout symptoms, but base their self-worth on their achievements and are workaholic, show a higher level of symptoms of these three personality disorders. Discriminant analysis also confirmed the hypothesis that all three borderline personality disorders, with performance-based self-esteem and workaholism as covariates, are relevant predictors of burnout, the most powerful among them being borderline personality disorder. Keywords: burnout, performance-based self-esteem, workaholism, personality disorders ::INTRODUCTION In line with Maslach ( 1976), the majority of researchers perceive burnout primarily as the result of excessive pressure and stress at work, and many have examined the role of individual personality traits. Even though the findings uniformly confirm a higher degree of neuroticism in burned-out individuals, only a few authors have sought the causes for burnout primarily in pathological personality structure. This despite the fact that Freudenberger ( 1974), who first described this concept in specialist literature, already drew attention to the connection with narcissist personality disorder, and Cooper ( 1986) described the narcissist and masochist types of burnout in psychotherapists. Due to the high incidence of burnout, psychotherapists often encounter patients suffering from these problems. In their psychotherapy practice, the authors of this paper observe that the diagnostic assessment of the personality structure in the majority of patients with burnout syndrome shows the char - acteristics of middle-level borderline personality organization, relatively closer to a neurotic structure (Kernberg & Caligor, 2005), or moderately integrated (borderline) personality structure (PDM Task Force, 2006). Impairments are expressed primarily with regard to one’s self-image (identity or self) and self- direction (of emotions and behavior), which corresponds to the first of the 4 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat two common features of any borderline personality disorder as defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5, 2013). We thus sought to empirically explore these clinical observations through research. ::Performance-based self-esteem as an identity impairment The sense of self-worth (self-esteem and self-respect) entails a positive or a negative view of oneself or an individual’s global assessment of himself (Rosenberg, 1965) that measures the extent to which an individual believes he is capable, important, successful, and worthy of respect (Coopersmith, 1967). How positive and stable one’s self-esteem is also depends on what an individual builds it on. If he builds it on perceived success or failure in an area that his self-esteem is based on (Crocker, 2002; Leary et al., 2003; Park, Crocker, & Mickelson, 2004), this is referred to by researchers as contingent self-esteem. This type of self-esteem is unstable because it is contingent on external sources (e.g., validation from others and competing with others) and is positively correlated with a high level of neuroticism and anxiety (Kernis, Cornell, Sun, & Berry, 1993; Roberts & Kassel, 1997; Judge, Erez, Bono, & Thoresen, 2002; Watson, Suls, & Haig, 2002), and with a low global self-esteem on the Rosenberg scale (Crocker, Luhtanen, Cooper, & Bouvrette, 2003). Research shows that performance-based self-esteem is the central part of a burned-out individual’s self-image (Hallsten, Josephson, & Torgén, 2005). It is related to (work) achievements and depends on external validation, which is why it is unstable. It has a strong to medium-strong positive correlation with burnout syndrome (Hallsten L., 2005; Dahlin, Joneborg, & Runeson, 2007; Pšeničny, 2009) and is the factor that relates stressors to burnout (Blom, 2012). Among the personality traits, performance-based self-esteem is the strongest predictor of burnout (Pšeničny, 2009) and is strongly correlated with workaholism. The need for positive self-esteem is a strong motivational factor leading and directing people’s experiences and functioning, with everyone seeking to maintain their level of self-respect within a given range (Rosen, 1991). When self-esteem is distinctly low or unstable, this need can be so strong that it leads to chronic over-engagement (both emotional and performance-based). Therefore, we believe that contingent self-esteem is about one’s self being split into a devalued true self and an almighty false self (Winnicot, 1980), which is why a person with such problems must constantly strive to overcome his low self-esteem and mask it with achievements. Thus the emotional or perfor - mance-based over-engagement observed in burned-out individuals involves 5 can burnout result from personality disorders? forced or compulsive efforts to maintain the almighty false part of one’s self that the individual is forced into by internalized aggressive introjects (Klein & Riviere, 2008). Hence, external circumstances are primarily the triggers of over-engagement rather than its cause. ::Workaholism as a self-direction impairment Workaholism means that a person works extremely hard due to a strong, uncontrolled internal urge (McMillan, O’Driscoll, & Burke, 2003). A worka - holic is addicted to work, being led by introjected, controlled motivation (Van den Broeck et al. , 2011; Van Beek, Hu, Schaufeli, Taris, & Schreurs, 2012). By working hard, he seeks to obtain validation from others, raise his self-esteem, or avoid shame, guilt, and isolation—that is, to satisfy the needs of his ego (Ryan & Deci, 2000). In our opinion, workaholism is also a behavioral expression of over-engagement intended to enhance an unstable self-image resulting from performance-based self-esteem (Pšeničny, 2009). It is a reflection of deficient (forced) self-direction of behavior and poor self-direction of feelings and emotions (DSM- 5, 2013) because the workaholic uses over-engagement to reduce the feelings of fear and anxiety he is overwhelmed by due to his low self-esteem, which may lead to burnout (Schaufelli, Taris, & van Rhenen, 2008; Pšeničny, 2009; Bak- ker, Demerouti, & Sanz-Vergel, 2014). Through workaholism, perfectionism (Taris, van Beek, & Schaufeli, 2010) and especially perfectionistic concerns (Jowett, 2014; Hill & Curran, 2015) are strongly associated with burnout. Accordingly, burnout can be conceived of as a process resulting from unsuccessful (workaholic) efforts to maintain a stable self-image through achievements, which are a component part of one’s self-esteem. Burnout is a decompensation, which is the final result of the process of self-exhaustion through overcompensated activity (Pšeničny, 2009). ::Borderline personality disorders and burnout Otto Kernberg uses the term “borderline personality organization” to refer to a specific pathological personality structure resulting from specific develop - mental ego deficits (cited in Praper, 1999). It involves a series of pathological personality traits that are so intense that they lead to significant functional impairments in the psyche and interpersonal relationships. Borderline per - sonality organization forms an internal structural basis for various clinical pictures or syndromes of borderline personality disorders (e.g., borderline, schizoid, and narcissistic). According to the DSM- 5, what all personality 6 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat disorders have in common are significant impairments in self (identity or self-direction) and interpersonal (empathy or intimacy) functioning (DSM- 5, 2013). Impairments in personality functioning are relatively stable across time and consistent across situations. Even though problems with identity (performance-based self-esteem) and self-direction (introjected, controlled motivation) manifested in workaholism and perfectionism draw attention to the possible specific vulnerability of bor - derline personality structure in burned-out individuals, only a few researchers have studied the links between borderline personality disorders and burnout. The majority establish a correlation between narcissism and burnout. In a psychiatric examination, most family physicians and pediatricians with burnout symptoms were diagnosed with narcissist personality disorder (Tizón, Gràcia, Larripa, Artigue, & Casajuana, 2013). Clergy burnout results from a narcissistic craving for admiration and appreciation, and the developmental needs of the clergy’s own family (Olsen & Grosch, 1991). Another study draws attention to the double role of narcissistic vulnerability, reporting that burnout symptoms primarily occur in more narcissistic employees that experience professional failure (Tecedeiro, 2010). Unresolved narcissistic issues may even contribute to psychotherapist burnout (Glickauf-Hughes & Mehlman, 1995). In addition to narcissistic personality disorder, other studies also establish features of “personality vulnerability” (Tillett, 2003) and borderline personality disorder (Alemany Martinez, Berini Aytés, & Gay Escoda, 2008) in burned- out individuals. We believe that, due to the deficits described above that are common to all borderline personality disorders, any of them can contribute to burnout; narcissistic personality disorder may be the most obvious one primarily because it has been predominating recently. Only a few studies of this type have been carried out to date, and we be - lieve that this possible aspect of burnout causes should also be studied more systematically, ultimately also because this syndrome has become more com - mon over the past decades and because simultaneously therapists have been reporting an increasingly predominant number of patients suffering from borderline personality disorders. These two trends may be related, especially because people with a borderline personality disorder experience life situa - tions and interpersonal relationships as extremely stressful (Powers, Gleason, & Oltmanns, 2013). ::Research issue We examined whether burnout is correlated with borderline, narcissistic, and schizoid borderline personality disorder. In addition, we explored whether 7 can burnout result from personality disorders? burnout is associated with these borderline personality disorders through performance-based self-esteem and workaholism. ::Hypotheses This study proceeds from the premise that performance-based self-esteem (evaluating oneself by achievements) reflects a weak identity and that com - pulsive hard work or workaholism results from introjected motivation (i.e., weaker self-direction), which are the essential general criteria defined by the DSM-5 (2013) for any borderline personality disorder. Accordingly, we ex - pected that the majority of individuals showing more symptoms of borderline, schizoid, and narcissistic borderline disorder would also show more symptoms of performance-based self-esteem and workaholism, and ultimately burnout as well. These assumptions can be used to define the working hypotheses. H 1 : The average number of performance-based self-esteem, workaholism, and burnout symptoms suitably discriminates between groups of in- dividuals at risk for burnout (dependent variable: relaxed, challenged, worn out, burned out). H 2 : H 2a : The average number of symptoms of the three border- line personality disorders (borderline, schizoid, and narcis- sistic) differs significantly by four burnout risk groups (inde- pendent variable: relaxed, challenged, worn out, burned out). H 2b : Significant correlation is expected between the categories of individu- als with low or high level of borderline personality disorders (borderline, schizoid, and narcissistic) and the burnout risk categories (relaxed, chal- lenged, worn out, and burned out). H 3 : The set of borderline personality disorders (borderline, schizoid, and narcissistic) suitably discriminates between various burnout classes, taking into account performance-based self-esteem and workaholism as covariates. ::METHODS ::Respondents and instruments Respondents: The random sample included respondents that completed the questionnaires posted on the Human Resources Development Institute’s 8 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat website from July 2013 to August 2015. A total of 3,393 individuals ( 2,461 women and 932 men) completed the questionnaires. The youngest respondent was 15 years old and the oldest was 80 years old ( M = 35.99, SD = 10.917). Instruments: The Adrenal Burnout Syndrome Questionnaire or ABSQ (Pšeničny, 2007) is composed of four scales: body symptoms ( 45 items), emotional symptoms ( 94 items), behavioral symptoms ( 61 items), and cognitive symptoms ( 46 items). Cronbach’s a = 0.981. The outcomes are the variable “average adrenal burnout syndrome rate” or ABSRa (with scores from 0 to 3), hereinafter referred to as “burnout,” and the categorical variable “burnout classes” or ABSCl ( 0 = no symptoms; 1 = mild; 2 = medium; 3 = strong burnout). The latter was further converted into the dichotomous variable “burnout categories” or ABSCat (low level = no symptoms; high level = 1, 2, 3). The respondents were then further divided into four burnout risk groups, whereby the variable “risk” (ABSRi) was obtained. The classification procedure is described below under “H 1 testing.” The Performance-Based Self-Esteem Scale or PBSE Scale (Hallsten, Jo - sephson, & T orgén, 2005) with four statements measures performance-based self-esteem. The responses were rated on a five-point Likert scale ( 1 = Fully disagree, 2 = Somewhat disagree, 3 = Neither agree nor disagree, 4 = Somewhat agree, 5 = Fully agree). With the author’s permission, the questionnaire was translated into Slovenian by Andreja Pšeničny and Mitja Perat, after which it was back-translated into English by an English specialist. The questionnaire’s reliability measured using Cronbach’s alpha was 0.863. The outcome is the variable “average test score” or, as we called it, “performance-based self-esteem” or PBS (with scores from 1 to 5). This variable was further converted into the dichotomous variable “performance-based self-esteem categories” or PBSCat (low level = 1–2.45; high level = 2.46–5). The criterion used was the average score reported by the authors of this scale. The Work Addiction Risk Test or WA RT (Robinson, 2014) comprises 25 items and measures the tendency for compulsive hard work or workaholism. With the author’s permission, the test was translated into Slovenian by Andreja Pšeničny and Mitja Perat, after which it was back-translated into English by an English specialist. Cronbach’s a = 0.947. The responses were rated on a four-point Likert scale ( 1 = Never true, 2 = Sometimes true, 3 = Often true, 4 = Always true). The outcome of the test is the variable “workaholism” or WORKHL (with scores ranging from 25 to 100). The average score reported 9 can burnout result from personality disorders? by Robinson is 47, which was used as the criterion for converting this variable into the dichotomous variable “workaholism categories” or WORKCat (low level: 25–47; high level = 48–100). The screening tests for borderline (BORDQ), schizoid (SCHIZQ), and narcissistic (NARCQ) borderline personality disorder were designed by the authors based on the criteria for diagnosing these personality disorders as recommended by the DSM- 5 (DSM-5, 2013). Each test includes 40 items with described symptoms, to which the respondents reply with Yes or No. The reliability of these tests was verified using Cronbach’s a (BORDQ = 0.898; SCHIZQ = 0.868; NARCQ = 0.868). We decided to design these screening tests because clinical questionnaires are not allowed to be posted on the in - ternet. The outcomes are the variables “borderline bpd” (BORD), “schizoid bpd” (SCHIZ), and “narcissistic bpd” (NARC), with scores ranging from 0 to 40. Half of the symptoms (low level = 0–20; high level = 21–40) was used as the criterion for classifying these variables into the dichotomous variables “borderline bpd categories” (BORDCat), “schizoid bpd categories” (SCHIZ - Cat), and “narcissistic bpd categories” (NARCCat). When individuals with a high level of symptoms of at least one of the bor - derline personality disorders ( 60%) were separated from those not showing these symptoms ( 40%), the variable “borderline disorder categories” (BPDCat) was obtained. Part of the respondents showed a high level of symptoms for more than one borderline personality disorder (the comorbidity of borderline personality disorders), which is a common feature (Lenzenweger, Lane, Lo - ranger, & Kessler, 2007). Therefore, we defined the variable “bpd comorbid - ity” (BPDCOM) to classify respondents by number of comorbid borderline personality disorders. Forty percent of the respondents did not show a high level of borderline personality disorder symptoms, 25.5% of them showed a high level of symptoms of one borderline personality disorder, 21.3% showed symptoms of two, and 14.2% showed symptoms of all three borderline per - sonality disorders. The variables were divided into two groups: “BURNOUT” (ABSBR, PBS, and WORKHL) and “BPD” (BORD, SCHIZ, and NARC). 10 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat ::Results 0. Descriptive sample statistics Table 1: Descriptive statistics for the testing results (mean scores and standard deviations) M SD Burnout (ABSRa) 0.90 0.68 Performance-based self-esteem (PBS) 3.46 1.11 Workaholism (WORKHL) 61.26 17.42 Borderline bpd (BORD) 16.49 8.27 Schizoid bpd (SCHIZ) 17.72 7.46 Narcissistic bpd (NARC) 18.73 7.74 Because the result distributions had an inappropriate kurtosis and the distribution skewed left in the case of ABSRa, resulting in the p-value of the Kolmogorov–Smirnov normality test being below 0.05, a two-step transfor - mation in SPSS (Templeton, 2011) was carried out. We used this approach because it makes it possible to retain the basic parameters ( M and SD). The transformations normalized the distributions ( p > 0.05). The differences in the test scores between the low- and high-level categories (see the Methods section for the variables used) are shown in Table 2. Table 2: Frequencies, descriptive statistics, and differences in test scores by category “low level” and “high level” for burnout, performance-based self- esteem, workaholism, and borderline, schizoid, and narcissistic borderline disorder n M SD t t-test df p ABSCat low 1,054 0.13 0.33 −78.99 2,844.59 0.00 ** high 2,349 1.25 0.48 PBSCat low 699 1.88 0.47 −82.72 1,957.63 0.00 ** high 2,694 3.86 0.83 WORKCat low 1,360 44.39 9.49 −78.27 3,226.72 0.00 ** high 2,033 72.54 11.34 BORDCat low 1,115 12.02 5.64 −77.13 2,772.42 0.00 ** high 2,278 25.62 4.37 11 can burnout result from personality disorders? SCHIZCat low 2,158 13.31 4.95 −77.12 2,995.12 0.00 ** high 1,235 25.42 4.05 NARCCat low 2,019 13.68 4.99 −77.31 3,193.95 0.00 ** high 1,374 26.15 4.34 ** Statistically significant difference p < 0.01 Notes: ABSCat = adrenal burnout syndrome categories; PBSCat = performance-based self-esteem categories; WORKCat = workaholism categories; BORDCat = borderline bpd categories; SCHIZCat = schizoid bpd cat - egories; NARCCat = narcissistic bpd categories 1. H 1 testing: In order to determine whether performance-based self-esteem and worka - holism are factors that indicate the presence of borderline personality disorders in burned-out individuals, we first had to divide the respondents into suitable groups that differed significantly by these three variables. Similar to Hallsten et al. (Hallsten, Josephson, & Torgén, 2005), we divided the respondents into four groups according to the burnout risk they were exposed to: relaxed, challenged, worn out, and burned out . The difference is that Hallsten et al. used low and high levels of performance-based self-esteem ( PBSCat) and low and high levels of burnout ( ABSCat) as their classification criteria, whereas we also added a third one: a low and high level of workaholism ( WORKCat) because both of these variables are statistically positively correlated with burn - out ( PBS-ABSRa: r = 0.54; p < 0.01; WORKHL-ABSRa: r = 0.46; p < 0.01). In this way we obtained the variable “risk” (ABSRi) with four groups of respondents with a different level of burnout risk: relaxed (low level of performance-based self-esteem, workaholism, and burnout), challenged (high level of performance-based self-esteem and/or workaholism and/or low level of burnout), worn out (low level of performance-based self-esteem and workaholism, and high level of burnout), and burned out (high level of performance-based self-esteem and/or workaholism and burnout). This classification was then tested through a discriminant analysis. The classification variable used was “burnout risk” or ABSRi (relaxed, challenged, worn out, and burned out), and the independent variables used were “BURN - OUT” (ABSRa, PBS, and WORKHL). 12 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat Table 3: Descriptive statistics (mean scores and standard deviations) for performance-based self-esteem, workaholism, and burnout by burnout risk groups, and results of a one-way analysis of variance M SD F df(3.3389) p Performance-based self-esteem Relaxed 1.76 0.59 2,769.18 0.00** Challenged 4.13 0.62 Worn out 2.14 0.59 Burned out 4.07 0.65 Workaholism Relaxed 38.65 10.68 776.99 0.00** Challenged 69.85 15.77 Worn out 50.28 11.94 Burned out 67.31 13.85 Burnout Relaxed 0.13 0.31 1,429.859 0.00** Challenged 0.36 0.23 Worn out 0.83 0.42 Burned out 1.39 0.47 ** Statistically significant difference p < 0.01 Table 4: Testing the equality of centroids for burnout risk (ABSRi) groups Functions tested Wilks’s lambda χ 2 df p 1 / 3 0.28 4,333.28 9 0.00 ** 2 / 3 0.69 1,281.82 4 0.00 ** 3 0.99 5.63 1 0.02 * ** Statistically significant difference p < 0.01 * Statistically significant difference p < 0.05 Table 5: Relative influence of predictors on functions’ values (standardized coefficients) and correlation between variables and functions (structural matrix) Standardized coefficients Functions Structural matrix Functions 1 2 3 1 2 3 Burnout rate 0.75 −0.73 −0.05 0.89 * −0.45 0.00 Performance-based self-esteem 0.34 0.68 −0.84 0.62 0.68* −0.39 Workaholism 0.20 0.41 1.04 0.55 0.52 0.65 * * Maximum absolute correlations between each variable and individual discriminant function 13 can burnout result from personality disorders? Table 6: Function values by centroid ABSRi Function 1 2 3 Relaxed −2.51 −0.43 −0.06 Challenged −1.21 1.55 0.05 Worn out −0.90 −1.52 −0.12 Burned out 0.77 0.24 0.01 Note: ABSRi = burnout risk groups All three functions are statistically significant (Table 4). The first function has the most significant positive statistical correlation with burnout, the second with performance-based-self-esteem, and the third with workaholism (Table 5). The value of functions by centroid and a one-way analysis of variance accurately show how these four groups differ from one another in terms of performance-based self-esteem, workaholism, and burnout. The relaxed have the lowest level of performance-based self-esteem, workaholism, and burnout. The challenged show the highest level of workaholism (post hoc p < 0.01), their performance-based self-esteem is higher than that of the first two groups (post hoc p < 0.01) and the same as that of the burned-out group (post hoc p > 0.05), and their burnout is slightly higher than that of the relaxed, but considerably lower than that of the worn out and burned out (post hoc p < 0.01). The worn out have a slightly higher performance-based self-esteem than the relaxed, but significantly lower than the challenged and burned out (post hoc p < 0.01); their level of workaholism is higher than that of the relaxed, but notably lower than that of the challenged and burned out (post hoc p < 0.01). The burned out show the highest level of performance-based self-esteem (just like the challenged), the highest level of workaholism (post hoc p < 0.01), and the highest level of burnout (post hoc p < 0.01; Tables 3 and 6). Originally, 78.6% of respondents were suitably classified into burnout risk groups according to the predictors “performance-based self-esteem,” “worka - holism,” and “burnout rate.” Based on the discriminant analysis results, a final classification of respondents by the variable ABSRi was carried out (Table 7). 14 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat Table 7: Classifying respondents into groups by burnout risk rate (the variable ABSRi) following discriminant analysis Low ABSCat High ABSCat Low PBSCat and WORKCat Relaxed 490 (14.4%) Worn out 499 (14.7 %) High PBSCat and WORKCat Challenged 745 (22%) Burned out 1,659 (48.9%) Note: ABSCat = adrenal burnout syndrome categories; PBSCat= performance-based self-esteem categories; WORKCat = workaholism categories The first hypothesis can be confirmed. Burnout risk groups (relaxed, chal - lenged, worn out, and burned out) in the new variable ABSRi differ significantly from one another by all three variables (i.e., performance-based self-esteem, workaholism, and burnout). 2. H 2 testing: If our hypothesis that performance-based self-esteem and workaholism predict the presence of borderline, schizoid, and narcissistic borderline per - sonality disorder is correct, the four respondent categories should also differ by the average number of symptoms of borderline personality disorders. The relaxed and worn-out individuals should have the smallest number of symp - toms (even though the latter show symptoms of psychophysical exhaustion), and the challenged and burned out should have significantly more symptoms, Figure 1: Comparison of the average number of symptoms of borderline, schizoid, and narcissistic borderline personality disorder between burnout risk groups (relaxed, challenged, worn out, and burned out) 15 can burnout result from personality disorders? even though the challenged do not show any burnout signs. To check this, we carried out a repeated measures analysis of variance. Because Mauchly’s test showed that sphericity ( χ 2 (2) = 189.60, p < 0.01) was violated, we corrected the degrees of freedom using Greenhouse–Geisser sphericity estimates. The results confirmed our assumption. Table 8: Summary of one-way repeated measures analysis of variance: borderline personality disorder symptoms (“BPD” variables) by relaxed, challenged, worn-out, and burned-out group (ABSRi variable) Variability source SS df MS F p Between individuals Cross section 133,682.57 3 44,560.86 521.37 0.00** Error 289,652.90 3,389 27.92   Within individuals BPD 6,515.56 1.90 3,450.75 111.68 0.00** BPD*ABSRi 2,814.11 5.67 496.80 16.08 0.00** Error 197,721.05 6,398.99 30.90 ** Statistically significant difference p < 0.01 The challenged and burned-out individuals have twice as many symptoms of all three borderline personality disorders as the relaxed, and also half again as many as the worn out; the differences are statistically significant between and within individuals (Figure 1, Table 8). A post-hoc comparison of differ - ence also shows no statistically significant differences between the challenged and the burned out in the average number of symptoms for borderline ( t = 0.38; p = 0.70) and schizoid ( t = 0.37; p = 0.71) borderline personality disorder, whereas the symptoms of narcissistic borderline personality disorder among the challenged are significantly higher than in the burned out ( t = 3.47, p = 0.001). The first part of the hypothesis—that performance-based self-esteem and workaholism in burnout indicate the presence of borderline personality disorders—can therefore be confirmed. If it is primarily individuals with borderline personality disorders that burn out, then one would expect a considerably higher percentage of those with a high level of borderline personality disorders among the challenged and burned out than among the relaxed and worn out. 16 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat Table 9: Comparison of the percentage of respondents with a high and low level of borderline personality disorder symptoms (BPDCat variable) between burnout risk groups (ABSRi variable) Borderline disorder category Low High Burnout risk group Relaxed n 429 61 % 87.6% 12.4% Challenged n 187 558 % 25.1% 74.9% Worn out n 312 187 % 62.5% 37.5% Burned out n 429 1,230 % 25.9% 74.1% Note: 0 cells ( 0.0%) have an expected number below 5. The minimum expected number is 195.97. As expected, three-quarters of the challenged and burned out also show a high level of symptoms for at least one borderline personality disorder, whereas the corresponding share among the relaxed is only slightly above one-tenth, and among the worn out just over one-third (Table 9). The differences are statistically significant ( χ 2 (3) = 774.31, p < 0.01). Hence, the second part of the hypothesis can also be confirmed. 3. H 3 testing: In the last stage, we were interested in whether borderline, schizoid, and narcissistic borderline disorder is a suitable predictor of burnout, taking into account performance-based self-esteem and workaholism as covariates. We expected that these variables would make possible classification into four burnout classes: no symptoms, and mild, medium, and strong burnout. Table 10: Testing the equality of centroids by burnout class (ABSCl) Functions tested Wilks’s lambda χ 2 df p 1 / 3 0.77 696.36 9 0.00** 2 / 3 0.99 5.11 4 0.28 3 1.00 0.33 1 0.57 ** Statistically significant difference p < 0.01 * Statistically significant difference p < 0.05 17 can burnout result from personality disorders? Table 11: Relative influence of predictors on the functions’ values (standardized coefficients) and the correlation between the “BPD” variables and functions (structural matrix) Standardized coefficients Functions Structural matrix Functions 1 2 3 1 2 3 Borderline BPD 0.56 −1.19 −0.31 0.92* −0.38 −0.14 Schizoid BPD 0.33 0.44 1.02 0.79* 0.47 −0.45 Narcissistic BPD 0.34 1.02 −0.59 0.72* 0.16 0.68 * Maximum absolute correlations between each variable and individual discriminant function Table 12: Function values by centroid Burnout class Function 1 2 3 No symptoms −0.79 −0.03 0.01 Mild 0.06 0.00 −0.02 Medium 0.24 0.05 0.01 Strong 0.99 −0.09 0.01 Note: mild = burnout class 1; medium = burnout class 2; strong = burnout class 3. Only the first function is statistically significant (Table 10). All three border - line personality disorders are strongly correlated with it, but most strongly the borderline version, which is the strongest predictor (Table 11). The function values by centroid show that respondents with no burnout symptoms have only a few borderline personality disorder symptoms and that the intensity of burnout by class increases with the growth of symptoms for all three bor - derline personality disorders (Table 12). ::DISCUSSION The motive for this research was psychotherapy experience with burned-out patients, for the majority of whom the diagnostic procedure showed certain early development impairments typical of people suffering from borderline personality disorders. These impairments are usually moderate and can there - fore be overlooked, and consequently excessive hard work and perfectionism can be ascribed to an anankastic personality structure. However, in doing so, the key differentiating factor can be overlooked—that is, that controlled motivation that ultimately leads to exhaustion originates from a deficit in 18 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat self-esteem that is unstable and dependent on achievements and external validation (i.e., performance-based self-esteem) (Hallsten, Josephson, & Torgén, 2005; Pšeničny, 2009). Without this, self-esteem is undermined and therefore individuals with performance-based self-esteem try hard through performance-based (and emotional) engagement to maintain a positive self- image. This engagement exceeds all limits because hard work is replaced by its compulsive form or workaholism (overcompensated pseudo-practice), which finally threatens psychophysical health; hence, this over-engagement is also an indicator of poorer self-direction. These are the two key criteria of borderline personality disorders (DSM- 5, 2013). There are few studies in the literature seeking a link between borderline personality disorders and burnout; they all associate this syndrome with narcissism, and only one study links it to borderline personality disorder. In their psychotherapy practice, the authors have established that not only narcissistic, but also other types of personality disorders occur in burned-out individuals, and therefore they investigated whether burnout is associated with borderline, narcissistic, and schizoid borderline personality disorder, and whether performance-based self-esteem as an identity impairment and worka - holism as a self-direction impairment are also indicators of these disorders. The survey was carried out on a random sample of 3,393 individuals, using the following six questionnaires: the Adrenal Burnout Syndrome Question - naire (Pšeničny, 2007), the Performance-Based Self-Esteem Scale (Hallsten, Josephson, & Torgén, 2005), the Work Addiction Risk Test (McMillan, O’Driscoll, & Burke, 2003), and the screening tests for borderline, schizoid, and narcissistic personality disorder designed by the authors of this paper. The authors expected that individuals that showed more symptoms of borderline, schizoid, and narcissistic personality disorder would also base their self-worth more on their achievements and would be greater workahol - ics. Because burnout results from psychological and physical self-exhaustion through excessive engagement, which also manifests itself as workaholism (Pšeničny, 2009), a similar connection with borderline personality disorders was also expected with regard to burnout. T o this end, the variables “performance-based self-esteem,” “workaholism,” and “burnout” were first converted into a new variable (“burnout risk”) that differentiates between the respondents by these three criteria. Similar to Hall - sten et al. (Hallsten, Josephson, & Torgén, 2005), respondents were divided into four groups. The first group, or the “relaxed,” do not base their self-worth on achievements, are not workaholics, and are not burned out. The second group, or the “challenged,” are those that are not (yet) burned out, but are workaholics and base their self-worth on achievements. The “worn-out” do 19 can burnout result from personality disorders? not show any symptoms of performance-based self-esteem and workaholism, and their burnout level is at nearly half the level as that among the “burned out,” who, just like the challenged, base their self-worth on achievements and are workaholics. In line with the hypothesis that performance-based self-esteem and worka - holism indicate the presence of symptoms of borderline personality disorders, the authors accordingly expected that the average number of symptoms of these disorders among the challenged and the burned out would be significantly higher than the relaxed and worn out, even though the latter show signs of burnout. A one-way repeated measures analysis of variance fully confirmed this assumption on the sample selected. However, it is interesting that there were no differences between the challenged and the burned out in the aver - age number of symptoms of borderline and schizoid personality disorder, whereas the challenged showed an even slightly higher number of symptoms of narcissistic personality disorder than the burned out. Accordingly, three-quarters of challenged and burned-out individuals also showed a higher level of symptoms of borderline personality disorders, whereas only one-tenth of these could be found among the relaxed and just over one-third among the worn out. The survey confirmed the authors’ empirical observations from their clini - cal psychotherapy practice that the majority of the burned-out individuals in the sample show symptoms of all three borderline personality disorders (i.e., borderline, schizoid, and narcissistic) and that performance-based self-esteem and workaholism also indicate the presence of these disorders. A higher level of narcissism among the challenged than among the burned out may be explained with Miguel Tecedeiro’s findings on the double role of narcissism (Tecedeiro, 2010): as long as narcissistic individuals experience professional success, they are positively motivated by narcissim, but when they experi - ence failure, the compulsive part comes to the fore and drives them into workaholism, with narcissistic vulnerability triggering burnout symptoms. The authors believe that, as an identity impairment, performance-based self- esteem simply denotes a specific vulnerability, whereas workaholism already indicates that self-esteem is being threatened and that the individual is trying to keep it positive at any cost. The average number of symptoms of borderline personality disorders among the challenged and the burned out accounts for approximately half of all the symptoms possible, which means that only a moderate, rather than extreme, number of symptoms is present among the majority of these respondents. This, however, agrees with the clinical observations because, in the authors’ psychotherapy practice, a moderately integrated borderline personality struc - 20 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat ture has been diagnosed in the majority of burned-out individuals (PDM Task Force, 2006). In conclusion, the authors were interested whether borderline, schizoid, and narcissistic borderline personality disorder is a suitable predictor of burnout, taking into account performance-based self-esteem and workaholism as co - variates. The discriminant analysis also confirmed this hypothesis: all three borderline personality disorders were suitable predictors and burnout increases with the number of borderline personality disorder symptoms. The authors were slightly surprised to see that the borderline personality disorder was the strongest predictor because the majority of sources primarily associate burnout with narcissism. However, taking into account that fear of loss (e.g., fear of losing one’s job or relationship) is often the strongest motivator in individu - als with a borderline personality disorder, one can understand why this fear can also easily trigger workaholism and consequently burnout. In addition, such chronic fear may also automatically lead to emotional exhaustion (Ma - slach, 1976) and loss of motivation (Pšeničny, 2009). This could also explain the chronic feeling of being overburdened regardless of the actual external burdens. In this case, emotional distancing of burned-out individuals may result from the defense mechanism of splitting—that is, devaluation due to the disappointment over the results of over-investment. The random (non-representative) sample is the main limitation of this study, preventing generalization of the results to the general population despite the large number of respondents included ( 3,393). In addition, one should bear in mind that on no account do the screening test results represent a clinical diagnosis of border personality disorder, but they can point to a specific vul - nerable individual. Therefore, the results indicating the presence of borderline personality disorders among burned-out individuals should be perceived primarily as a warning that this is highly likely. Hence, this study should be repeated on a representative sample and in - clude clinical borderline personality disorder tests. In any case, the results draw attention to the fact that the connection between borderline personality disorders and burnout, as well as performance-based self-esteem and worka - holism, should be studied in greater detail. Likewise, the relationship between the last two attributes should be examined more thoroughly and the ques - tion of whether burnout is also associated with other personality disorders in addition to those studied here should be explored. It would especially make sense to explore the possible connection between workaholism and obsessive- compulsive personality disorder. 21 can burnout result from personality disorders? ::REFERENCES Alemany Martinez, A., Berini Aytés, L., & Gay Escoda, C. (2008). The burnout syndrome and associated personality disturbances. The study in three graduate programs in Dentistry at the University of Barcelona. Medicina Oral Patología Oral y Cirurgía Bucal 13(7), 444-450. Bakker, A. B., Demerouti, E., & Sanz-Vergel, ,. A. (2014). Burnout and work engagement: The JD-R approach. Annual Review of Organizational Psychology and Organizational Behavior, 1: 389–411. Blom, V. (2012). Contingent self-esteem, stressors and burnout in working women and men. Work, 43(2), 123-131. Cooper , A. (1986). Some limitations on therapeutic effectiveness: The “burnout syndrome” in psychoanalysts. The Psychoanalytic Quarterly, 55(4), , 576-598. Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: Freeman. Crocker, J. (2002). The Costs of Seeking Self–Esteem. Journal of Social Issues, 58, 3, 597–615. Crocker, J., Luhtanen, R. K., Cooper, M. L., & Bouvrette, S. (2003). Contingencies of self-worth in college students: Theory and measurement. Journal of Personality and Social Psychology, 85, 894-908. Dahlin, M. E., Joneborg, N., & Runeson, B. (2007). (2007). Performance-based self-esteem and burnout in a cross-sectional study of medical students. Medical Teacher, 29 (1), 43-48. DSM-5. (2013). Diagnostic and Statistical Manual of Mental Disorders, 5. New York: American Psychiatric Association. Freudenberger, H. J. (1974). Staff Burnout. Journal of social Issues, 30, 159-165. doi: 10.1111/j.1540-4560.1974. tb 00706.x. Glickauf-Hughes , C., & Mehlman, E. (1995). Narcissistic issues in therapists: Diagnostic and treat - ment considerations. Psychotherapy: Theory, Research, Practice, Training, 32(2), 213-221. Hallsten, L. (2005). Burnout and wornout: Concepts and data from a national survey. In A. S. Antoniou, & C. L. Cooper , Research companion to organizational health psychology (pp. 516-537). Cheltenham, U. K.: Edward Elgar. Hallsten, L., Josephson, M., & Torgén, M. (2005). Performance-based self-esteem. A driving force in burnout proces and its assessment. Arbete och Halsa, 4, pp. 117–139. Hill, A., & Curran, T. (2015, julij 31). Multidimensional Perfectionism and Burnout. Retrieved from Personal - ity and Social Psychology Review: http://psr.sagepub.com/content/early/ 2015/07/25/1088868315596286. abstract. doi: 10.1177/1088868315596286 Jowett, G. E. (2014). Perfectionism, engagement and burnout in youth sport and dance: A self-determination theory perspective. PhD thesis. Leeds, UK: University of Leeds. Judge , T. A., Erez, A., Bono, J. E., & Thoresen, T. J. ( 2002). Are measures of self-esteem, neuroti - cism, locus of control and generalized self-efficacy indicators of a common core construct? Journal of Personality and Social Psychology, 83(3) , 693-710. Kernberg, O. F., & Caligor, E. (2005). A psychoanalytic theory of personality disorders. In M. F. Lenzenweger, & J. F. Clarkin , Major theories of personality disorder (pp. 114-156). New York: Guilford Press. Kernis, M. H., Cornell, D. P., Sun, C., & Berry, A. (1993). There’s more to self-esteem than whether it is high or low: The importance of stability of self-esteem . Journal of Personality and Social Psychology, 65(6) , 1190-1204. Klein, M., & Riviere, J. (2008). Ljubav, mržnja i reparacija. Novi Sad: Akademska knjiga. Leary, M. R., Gallagher, B., Fors, E., Buttenmre, N., Baldwin, E., Kennedy, K., & Mills, A. (2003). The invalidity of disclaimers about the effects of social feedback on self- esteem. Personality and Social Psychology Bulletin, 29, 623-636. Lenzenweger, M., Lane, M., Loranger, A., & Kessler, R. (2007). DSM-IV personality disorders in the National Comorbidity Survey Replication. Biological Psychiatry. 62(6), 553–564. doi: 10.1016/j. biopsych. 2006.09.019. Maslach, C. (1976). Burned-out. Human Behaviour, 5, 16-22. 22 Anthropos 1-2 (241-242) 2016, str. 1-23 andreja pšeničny, mitja perat McMillan, L., O’Driscoll, M., & Burke, R. (2003). Workaholism: a review Of theory, research and new directions. In C. Cooper , & I. Robertson , International review Of industrial and organiza- tional psychology (pp. 167—190). New York: John Willey. Olsen, D., & Grosch, W. (1991). Clergy Burnout: A Self Psychology and Systems Perspective. Journal of Pastoral Care & Counseling, 45 (3), 297-304. Park , L. E., Crocker, J., & Mickelson, K. D. (2004). Attachment styles and contingencies of self- worth . Personality and Social Psychology Bulletin, 30 , 1243-1254. PDM Task Force. (2006). Psychodynamic Diagnostic Manual. Silver Spring: MD: Alliance of Psy - choanalytic Organisations. Powers, A., Gleason, M., & Oltmanns, T. (2013). ymptoms of borderline personality disorder predict interpersonal (but not independent) stressful life events in a community sample of older adults. Journal of Abnormal Psychology. 122 (2), 469-474. http://dx.doi.org/ 10.1037/a0032363. Praper, P. (1999). Razvojno analitična psihoterapija, 2. razširjena izdaja [Developmental analytic psychotherapy, 2. extended edition]. Ljubljana: Inštitut za klinično psihologijo. Pšeničny, A. ( 2007). Razvoj vprašalnika sindroma adrenalne izgorelosti (SAI) in preverjanje izhodišča Recipročnega modela izgorelosti [Development of Adrenal Burnout Syndrom Questionnaire and Testing the Basis of Reciprocal Burnout Model]. Psihološka obzorja, pp. 47–82. Pšeničny, A. ( 2009). Recipročni model izgorelosti : povezava med interpersonalnimi in intrapersonalnimi dejavniki nastanka sindroma izgorelosti : doktorska disertacija (neobjavljena). Ljubljana: Univerza v Ljubljani, Filozofska fakulteta, Oddelek za psihologijo. Roberts , J. E., & Kassel, J. D. (1997). Labile self-esteem, life stress, and depressive symptoms: Pro - spective data testing a model of vulnerability . Cognitive Therapy and Research, 21 (5) , 569-589. Robinson, B. E. (2014). Chained to the Desk: A Guidebook for Workaholics, Their Partners and Children and the Clinicians Who Treat Them. (. New York: New York University Press. Rosen, l. (1991). Self-esteem as a factor in social and domestic violence. British Journal of Psychiatry, 158, 18-23. Rosenberg, M. (1965). Society and the adolescent self-image. Prmceton NJ: Princeton . Ryan, R., & Deci, E. (2000). Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being. American Psychologist, 55 (1), 68-78. Schaufelli, w. B., Taris, T. W., & van Rhenen, W. (2008). Workaholism, Burnout, and Work En - gagement: Three of a Kind or Three Different Kinds of Employee Well-Being? Applied Psycholog; An International Review, 57, 173-203. Taris, T. W., van Beek, I., & Schaufeli, W. B. (2010). Why do perfectionists have a higher burnout risk than others? The mediational effect of workaholism. Romanian Journal of Applied Psychol - ogy, 12. Romanian Journal of Applied Psychology, 12, 1-7. Tecedeiro, M. (2010). Estudo exploratörio sobre burnout numa amostra portuguesa: O narcisismo como variåvel preditora da sindrome de burnout. Ånålise Psicolögica 28 (2), 311-320. Templeton, G. F. (2011). A Two-Step Approach for Transforming Continuous Variables to Nor - mal: Implications and Recommendations for IS Research. Communications of the Association for Information Systems, Vol. 28, pp. Article 4, dostopno na: http://aisel.aisnet.org/cais/vol 28/iss 1/4. Tillett, R. (2003). The patient within — psychopathology in the helping professions. Advances in Psychiatric Treatment 9 (4), 272-279, doi: 10.1192/apt. 9.4.272 . Tizón, J., Gràcia, P., Larripa, A., Artigue, J., & Casajuana, J. (2013). Preventive care for burnout: General practitioner pilot program in Barcelona. Health, 5, 24-31. doi: 10.4236/health. 2013.56A2005. Van Beek, I., Hu, Q., Schaufeli, W. B., Taris, T., & Schreurs, B. (2012). For Fun, Love, or Money: What Drives Workaholic, Engaged, and Burned-Out Employees at Work? Applied Psychology, 61 (1), 30–55. Van den Broeck, A., Schreurs, B., De Witte, H., Vansteenkiste, M., Germeys, F., & Schaufeli, W. (2011). Understanding Workaholics’ Motivations: A Self-Determination Perspective. Applied Psychology, 60 (4), 600-621. 23 can burnout result from personality disorders? Watson, D., Suls, J., & Haig, J. (2002). Global self-esteem in relation to structural models of per - sonality and affectivity . Journal of Personality and Social Psychology, 83(1), 185-197. Winnicot, D. W. (1980). Dijete, obitelj i vanjski svijet. Zagreb: Naprijed.