Psychopathological Functioning Levels (PFLs) and their possible relevance in psychiatric treatments: a qualitative research project

Qualitative research aims to broaden the understanding of clinical experiences and phenomena as they are lived through situations, rather than testing hypothesized relationships or causal explanations, that benefit best from quantitative methods [18, 19]. This central purpose is common to different approaches that have been developed in qualitative research. Nevertheless, each of them have its own traditions and methods, and a specification of the researchers’ conceptual frameworks is therefore required [18, 19] and guidelines have been developed in this regard [20].

In order to conveniently assess and present the results of our research, the following Qualitative Research Criteria (QRC) have been taken into account:

They were adopted with reference to the “Evolving guidelines for publication of qualitative research studies in psychology and related fields” – Table 1/B: “Publishability Guidelines Especially Pertinent to Qualitative Research” [19]. In more detail, QRC1 refers to the criterion B2 of the aforementioned Guidelines, QRC2 to the criterion B1, QRC3 to the criterion B5, QRC4 to the criterion B6, QRC5 to the criterion B4, QRC6 to the criterion B7, and QRC7 to the criterion B3. Finally, also the Consolidated Criteria for Reporting Qualitative Research (COREQ) [20] checklist has been applied.

In line with the procedures used by OPD-2 and DSM-5 [2, 16], subsequent steps were taken to identify core psychopathological dysfunctions to be rated by the novel instrument and to refine the levels of impairment in order to be diagnosed by a specific rating scale.

From 2002 until 2014, 316 outpatients referring to the Department of Mental Health - ASL Turin 4, Italy, were administered semi-structured clinical interviews on at least two different occasions before starting treatments. Participants were 84 males (26.5 %) and 232 females (73.5 %) with a mean age of 37.9 ± 11.9 years. To be included in this study, patients had to be diagnosed with the following diagnoses (according to the DSM-5 [16]): Schizophrenic and Psychotic Disorders, Depressive Disorders, Anxiety Disorders, Obsessive-Compulsive Disorder, Post-Traumatic Stress Disorder, Somatic Symptoms Disorders, Eating Disorders and Personality Disorders. Six individuals (1.9 %) were diagnosed with Schizophrenic and Psychotic Disorders, 63 (20 %) with Depressive Disorders, 129 (40.8 %) with Anxiety Disorders, 13 (4.1 %) with Obsessive-Compulsive Disorder, 7 (2.3 %) Post-Traumatic Stress Disorder, 13 (4.1 %) Somatic Symptoms Disorders, 11 (3.4 %) Eating Disorders and 74 (23.4 %) with Personality Disorders.

The tool to be developed is intended to assess different levels of psychopathological functioning underlying a variety of psychiatric disorders (QRC1). To this end, it must be specified which psychopathology the Authors refer to (QRC2).

In order to fulfil this requirement, the PFL research specifically referred to the Vulnerability Events Personality - Psychopathological Model (VEP-PM), a specific bio-psycho-social model [21‐24] for the comprehension and treatment of psychiatric disorders. VEP-PM is rooted in phenomenology, psychodynamics, and neuroscience [25‐35]. More in detail, VEP-PM is grounded on Alfred Adler’s psychodynamic psychopathology [25, 36], which is consistent with Cloninger’s definition [37] of temperament and character dimensions and current approaches to psychopathology thus highlighting the developmental perspective in the understanding of the functional structures of the mind [38].

According to the VEP-PM, the following pathogenic factors are considered:

Factors a) and b) are combined to generate suffering (VEP-PM causative psychopathological factors), while factor (c) rises from the patient’s attempt to face it (VEP-PM compensatory psychopathological factor).

A focus group of 8 experienced clinicians and researchers (3 males and 5 females) working at the APPs-TR network (3 psychiatrists and 5 clinical psychologists) discussed and proposed criteria to identify those psychopathological dysfunctions which needed to be rated, also considering those targets potentially helpful in managing treatments across a wide range of psychic disorders. Also, data saturation was discussed.

Two Phenomenological Selection Criteria (PhSC) and four Etiopathogenetic Selection Criteria (EtSC) were chosen to highlight significant specific dysfunctional areas.

As a subsequent step, the consistency of the five-areas rating model with other models proposed in literature to evaluate specific dysfunctional aspects has been checked.

The identification of the aforementioned core psychopathological areas emerges as positively related to a number of current research findings.

Experts’ consensus has been reached on the evaluation of severity in order to create a solid dimensional system of psychopathology [51]. According to recent research [52], there is a correlation between both severity and disability of a mental disorder and the level of impairment of individuals’ psychopathological functioning. Relatedly, outcome is predicted in a more accurate way by severity than by a mere disorder classification [17, 53].

More specifically, prior research with reference to “primitive”, “intermediate”, and “more advanced” levels of dysfunctions that characterize anaclitic and self-definitional disorders indicated that patients’ response to treatment is consistent with their diverging Personality Organization [54].

According to the VEP-PM theoretical model, the APPs-TR network focus-group decided that different dysfunctional levels in these areas should consider Personality Organization in terms of mechanisms of defence, coping, and creative compensations. These mechanisms are overall considered to underlie the quality of patient’s attempts to face and reduce pain and discomfort.

Different dysfunctional degrees in core psychopathological dysfunctions were therefore considered only if meeting the following Etiopathogenetic Selection Criterion (EtSC).

(EtSC5) The levels of psychopathological dysfunctions relate to Personality Organization (“VEP PM compensatory psychopathological factor”), that is each impairment level should correspond to a different Personality Organization (PO).

PO refers to a set of enduring, mostly unconscious, brain mechanisms [12, 55] and psychological structures that dynamically organize mental processes and contents into a coherent organization. These dynamically organized structures and processes are assumed to be involved in self-structure, cognition, affect regulation, impulses, and quality of relationships [40, 56, 57], in turn determining the levels of impairment.

Nevertheless, PO is still a latent construct, which can only be inferred from manifest indicators [57].

Seven Psychopathological Functioning Levels (PFL) of each area met the required selection criterion.

They were ordered progressively [58] along a continuum [17] of pathological PO [59, 60], according to Kernberg’s model [61] and subsequent integrations provided by recent research focusing on defence mechanisms, sharing the assumption that levels of PO follow a developmental progression from severely undifferentiated and disorganized levels of PO to mature, integrated, and differentiated levels of PO [12].

In fact, the Borderline Personality Organization, which was originally divided in two levels [61], was classified in this study into three levels, according to prototypical descriptions of the Structured Interview of Personality Organization (STIPO) [Clarkin JF, Caligor E, Stern BL, Kernberg OF. Structured Interview of Personality Organization (STIPO). New York, unpublished document 2004] and to the Defence Mechanisms Rating Scale (DMRS) findings [62]: 1) with prevalent use of major image-distorting defences, 2) with prevalent use of minor image-distorting defences, 3) with prevalent use of both minor image-distorting and disawoval defences.

Additionally, a specific level was dedicated to the coexistence of Neurotic and Borderline Personality Organizations, namely when the use of the above mentioned immature defences is displayed only in situations characterized by severe threats and/or when it is widely intertwined with the use of neurotic defences (e.g., repression). In addition, this level significantly correlates with “Neurotic 2” PO, according to the STIPO [Clarkin JF, Caligor E, Stern BL, Kernberg OF. Structured Interview of Personality Organization (STIPO). New York, unpublished document 2004].

This seven-impairment levels rating model emerges as positively related to a number of current research findings.

Finally, the schedule was based on the following steps.

To date, a number of reliable and valid measures that assess personality functioning and psychopathology demonstrate that a self-other dimensional perspective has an empirical basis and significant clinical utility [81]. This approach was found to be informative in understanding those mental processes associated with psychopathology phenomenology, in planning treatment interventions, in anticipating treatment course and outcome, and in measuring change in treatment [81]. More in detail, current diagnostic and statistical research leads to consider that different degrees of maladaptive integration and balance of interpersonal relatedness and self-definition are involved across a variety of psychopathological pathways underlying symptoms [38]. These recent findings are also consistent with VEP-PM, strengthening Alfred Adler’s assumption concerning the related relevance of social interest and striving for power as effective factors of psychic health or disease. The patient’s emphasis on self-definition (internal perspective) or on relatedness (external perspective) delineates individual’s identity, meanings, and forms of cognition, feelings and affects, behaviors and reactions, qualities of relationships [38].

Patients with psychotic, borderline, neurotic, and mature Personality Organization were included and the PFL-RS provisional schedule overall provided 14 definitions for each area, in a two by two order on a progressive scale.

The semantic comprehension of the PFL-RS provisional schedule was refined by asking clinicians to read and provide feedback on the proposed definitions of the psychopathological dysfunctions.

All 70 definitions making up the schedule were evaluated for their comprehensibility by 20 reviewers (7 males and 13 females): 5 expert clinicians (2 psychiatrists and 3 clinical psychologists), 10 mental health professionals (MHP) (3 psychiatrists, 2 childhood and adolescence psychiatrists, 2 clinical psychologists, 3 nurses) and 5 trainee graduate psychologists. The task was carried out in two subsequent stages.

After the first reading, 20 corrections have been proposed by reviewers concerning ID area, 13 concerning CO area, 15 concerning NE area, 13 concerning AR area, and 5 concerning SS area. Furthermore, 7 of overall corrections were referring to dysfunctional level 1, 5 to dysfunctional level 2, 3 to dysfunctional level 3, 4 to dysfunctional level 4, 11 to dysfunctional level 5, 10 to dysfunctional level 6, and 1 to dysfunctional level 7.

These corrections were recognized as relevant and accepted by PFL-RTC scientific coordinator, supervisor and secretaries.

A second reading of the derivative revised version of the schedule was provided by reviewers. They added further corrections: 2 were concerning ID area, 1 was concerning CO area, 2 were concerning NE area, 4 were concerning SS area since no additional correction were proposed concerning AR area. Concerning dysfunctional levels, 1 correction was concerning level 3, 7 were concerning level 4, and 1 was concerning level 6, while no additional correction were proposed concerning levels 1,2 and 5. Also these corrections were afterwards implemented into the schedule.

For further details see Table 1.

As shown in Tables 2, 3, 4, 5 and 6, when PFL-RS definitions were considered satisfactory, the Italian schedule of Psychopathological Functioning Levels – Rating Scale (PFL-RS) was finalized [82]. Subsequently, an English translation of PFL-RS schedule was provided by a mother-language psychiatrist and psychotherapist; a back translation was then conducted to confirm the accuracy of the original text (Tables 7, 8, 9, 10 and 11). Patients were administered the Italian version of the PFL-RS as shown in Tables 2, 3, 4, 5 and 6.

PFL-RS can be repeated multiple times to assess patient’s psychopathological dysfunctions during the course of treatments.

Raters are asked to choose the definition that best describes patient’s worst psychic functioning at that moment. PFL rating may be different in each area of the same patient. Although short, a specific training is recommended in order to provide guidelines to conduct this assessment.

In summary, assessment procedures include a semi-structured interview (lasting approximately 1 h) and subsequent open interviews (lasting approximately 30 min), containing specific anamnestic questions and referring to clinical criteria.

More in detail, the interview is focused on recognizing different phases [2]: 1. Opening (symptoms, life events). 2. Dysfunctional repetitive patterns. 3. Self and others perception in different areas. 4. Past and present life style and life tasks. Unconscious processes can be inferred by contradictions and narrative structure.

Raters are also required to mark down the quality of patient-therapist relationship; this will be considered as a control variable during the evaluation, eventually suggesting absence of relationship, opposition versus willingness, idealized, dependent, supportive, fearful, preoccupate, dismissing, or cooperative relationship, only to name a few (see Tables 12 and 13).

In order to assess patients’ dysfunctions, more than a decade ago, the clinicians of the Mental Health Department of the ASL Turin 4 in Italy started using the “Primary Psychotherapeutic Focus Levels” (PPFLs) and “Personality Functioning Levels” (PFLs), two sets of definitions pioneering the PFL-RS (see Step G).

More in detail, from 2004 to 2007, when conducting the Brief-Adlerian Psychodynamic Psychotherapy (B-APP) therapists based their treatment strategies on the description of patient’s psychological functioning according to the PPFLs. Such a psychopathology-based psychotherapy showed efficacy and improved symptoms and global functioning in a sample of patients with Generalized Anxiety Disorder [83].

Furthermore, from 2006 to 2011, the multidisciplinary therapeutic team of Mental Health Service (MHS) in Chivasso (Turin, Italy) was trained and supervised to assess different PFLs which served as benchmark for the Supervised Team Management (STM) of clinical projects for patients with Borderline Personality Disorder in a community setting. Treatments included medications, unstructured psychological support (UPS) focused on socio-relational impairment or a specific time-limited psychotherapy (Sequential Brief-Adlerian Psychodynamic Psychotherapy – SB-APP), and rehabilitative interventions [22, 80]. A set of definitions of different impairment levels concerning the same five core areas that will be envisaged to be investigated by PFL-RS (see Table 14) was evaluated by two independent raters, in order to adjust the modulations of the therapeutic relationship to fit each specific patient [22, 23].

Overall, STM had to consider that patients, when at PFL-1 and PFL-2, unconsciously fear that their fragile identity might collapse (ID 1-2). In this regard, empathic validation and, to a lesser degree, clarification and affirmation could effectively convey constructive experiences opposite to precarious self-cohesion (ID 1-2). The latter is represented by patient’s inability to think when others are present (CO 1-2) and to tolerate (NE 1-2) the therapist as endowed of a separate existence (SS 1-2). At PFL-3, STM took into account that patients are largely engaged in denying self-other contradictory images (ID 3) as relevant for their life and behaviors (AR 3-4), since at PFL-4 this attitude only appears when facing specific tearing emotions and situations (ID 4). Treatments were aimed at reducing the sense of emptiness (NE 3-4; SS 3-4) and increasing continuity and adaptation: systematic consideration was paid to patients’ concrete way of thinking (CO 3-4), acting-out (AR 3-4), and discontinuities in their social relationships involving intimacy (SS 3-4).

In a preliminary clinical randomized study, thirty-five outpatients with BPD meeting inclusion criteria of service heavy users were evaluated. The study was aimed to evaluate the efficacy of PFLs informed clinical projects (STM), both including UPS and SB-APP, compared to Treatment As Usual (TAU) [80].

PFLs’ scoring was not included as patients’ assessment measure at baseline, since PFLs’ set of definitions had not been validated yet. However, it was found that overall PFL’s scoring distribution at baseline varied depending on each considered psychopathological area (see Table 15) and it did not overlap with the scores on Clinical Global Impression (CGI) [84], Symptom Checklist-90 Revised (SCL-90-R) [85] and Global Assessment of Functioning (GAF) [69], both in UPS and SB-APP group, thus providing specific additional clinical information (see Table 15). Compared to Treatment As Usual (TAU), after one year of treatment, STM showed an overall improvement in clinical severity (CGI) [84], global functioning (GAF) [69], and all nine psychopathological domains included in the diagnosis of BPD (CGI-M items 1–9) [80]. Both compliance to treatments and quality of therapeutic alliance were greater as well.

More in detail, one large (and also statistically significant) effect size was found for the outcome of identity disturbance (SMD 0.88, 95 % CI 0.16 to 1.60), a core dysfunctional area assessed by previous research [86].

This area was negatively correlated with self-harm and hospitalization rates. The reduction of dramatic occurrences and unscheduled interventions could highlight that a significant component of symptoms might be reduced by treatments informed by a specific assessment of underlying psychopathological dysfunctional areas [80].

Finally, the overall improvement in global functioning (GAF) [69] seems particularly important because patients’ quality of life remained poor, even after well-designed therapeutic interventions [80].

To date, PLS-RS was randomly administered to a small sample of 54 outpatients consecutively recruited at the Psychotherapy Unit in Settimo Torinese, Turin, Italy, together with the Temperament and Character Inventory (TCI) [37], Spielberger State and Trait Anxiety Inventory [87], Beck Depression Inventory [88], and the State-Trait Anger Expression Inventory [89]. Sixty % of the sample was composed by males and 40 % by females and their mean age was 29.03 ± 13.2 years (range: 17–64 years). The interviewers were either male (A.F.) or female researchers (S.F., E.C., B.S.) with MD or PhD credentials and extensive training (i.e., at least 5 years) in research. At the time of the study they were all working at the same facility (i.e., Psychotherapy Unit in Settimo Torinese, Turin, Italy) and all interviews were conducted there. Before starting this study, the researchers did not know the potential participants and contacted them using a face-to-face approach. Eight people refused to take part in this study because of lack of time. Before starting the interview, all participants were informed about the main goals of the study with the interviewers clarifying the main assumptions of this work. During the interview, only the researcher and the participant were present. Audio recording was not used because of research feasibility reasons; however the interviewer took notes of the interview during the assessment. Transcripts were not shared with participants and their feedback on the interview was encouraged but not formally required.

Three data coders coded the data according to the coding scheme provided by the authors. The Statistical Package for Social Sciences 21.0 (SPSS, SPSS Inc., Chicago, IL) was used for analysing the data. Concerning internal consistency, all PFL-RS subscales correlate strongly to each other. For example, Pearson’s correlation coefficient between internal and external perspectives of ID is r = .904, of CO is r = .901, NE is r = .922, AR is r = .614, SS is r = .,892. Moreover, HA on the TCI correlates positively with ID, CO, NE, and AR; finally, correlations have been also found between PFL-RS subscales and measures of depressive symptoms, anxiety, and anger. With more detail, ARe was inversely correlated with BDI (r = −.319, p = .024), STAI-trait (r = −.296, p = .035), and STAXI (r = −.396, p = .004). Also ARi was negatively correlated with STAI-trait (r = −.457, p = .001) and STAXI (r = −.431, p = .002). IDe was found to negatively correlate with STAXI (r = −.309, p = .029) as well as NE e (r = −.291, p = .040).

Finally, given the clinical use of this instrument considering each psychopathological areas and its global dimension, Cronbach’s alpha was .971. However, exhaustive research in order to assess PFL-RS validity, sensitivity and reliability is currently needed.