How paradoxical is ‘paradoxical’ outcome? Different pathways and implications

ABSTRACT

The present special issue includes single and multiple case studies which report on the phenomenon of ‘paradoxical’ outcomes in psychotherapy, operationalized as discordant outcome assessments within the same case. In an attempt to find ground for my published response, I endeavored to juxtaposition the cases from the different studies, and by doing that gain an overview of the contributions and arguments. As a result of that exercise, I suggest three different types of paradoxical outcome with different pathways and implications. Discussing the concept of ‘illusory mental health’ as a possible explanation for paradoxical outcome, I suggest that self-ratings within a nonclinical range, given by respondents assessed as psychologically distressed, could be perceived as ‘genuine’ markers of a tolerable level of coping and symptom relief, not only as an indication of a defensive attitude. I conclude that it might be more advantageous to relate ‘paradoxical’ outcomes to aspects of the therapeutic process, rather than to psychological dispositions of the clients.

ABSTRAKT

Die vorliegende Sonderausgabe enthält Einzel- und Mehrfachfallstudien, die über das Phänomen „paradoxer“ Ergebnisse in der Psychotherapie berichten und als nicht übereinstimmende Ergebnisbewertungen innerhalb desselben Falls operationalisiert werden. Um einen Grund für meine veröffentlichte Antwort zu finden, bemühte ich mich, die Fälle aus den verschiedenen Studien nebeneinander zu stellen und mir so einen Überblick über die Beiträge und Argumente zu verschaffen. Als Ergebnis dieser Übung schlage ich drei verschiedene Arten von paradoxen Ergebnissen mit unterschiedlichen Wegen und Implikationen vor. Ich diskutiere das Konzept der „illusorischen psychischen Gesundheit“ als mögliche Erklärung für das paradoxe Ergebnis und schlage vor, dass Selbsteinschätzungen innerhalb eines nichtklinischen Bereichs, die von Befragten als psychisch problematisch eingestuft werden, als „echte“ Marker für ein tolerierbares Bewältigungsniveau wahrgenommen werden könnten und Linderung der Symptome, nicht nur als Hinweis auf eine defensive Haltung sind. Ich komme zu dem Schluss, dass es möglicherweise vorteilhafter ist, „paradoxe“ Ergebnisse mit Aspekten des therapeutischen Prozesses in Beziehung zu setzen, als mit psychologischen Dispositionen der Klienten

RESUMEN

La actual cuestión específica incluye estudios de un caso único y múltiple que informan sobre el fenómeno de los resultados ‘paradójicos’ en la psicoterapia, los cuáles llevan a la práctica resultados discordantes en la evaluación de un mismo caso. En un intento de encontrar respuesta en el ámbito de la publicación, se realizó un gran esfuerzo para yuxtaponer los casos de los diferentes estudios, y al hacerlo obtener una visión general de las aportaciones y argumentos. Como resultado de ese ejercicio sugiero tres tipos diferentes de resultados paradójicos con diferentes vías e implicaciones. Al discutir el concepto de ‘salud mental ilusoria’ como una posible explicación para el resultado paradójico, sugiero que las evaluaciones dentro de un rango no clínico, dada por los encuestados evaluados con problemas psicológicos, podrían ser percibidas como marcadores ‘genuinos’ de un nivel tolerable de afrontamiento y alivio de los síntomas, no sólo como una indicación de una actitud defensiva. Concluyo que podría ser más ventajoso relacionar los resultados ‘paradójicos’ con aspectos del proceso terapéutico, en lugar de con las disposiciones psicológicas de los clientes.

ABSTRACT

Il presente numero speciale include studi di casi singoli e multipli che riportano il fenomeno degli esiti ‘paradossali’ in psicoterapia, operazionalizzati come discordanti procedure di assessment dei risultati all’interno dello stesso caso. Nel tentativo di trovare una base per la mia risposta e pubblicazione, ho cercato di giustapporre i casi dei diversi studi, ottenendo così una panoramica dei contributi e degli argomenti. Come risultato di ciò, suggerisco tre diversi tipi di risultati paradossali con percorsi e implicazioni differenti. Discutendo il concetto di `` salute mentale illusoria ‘‘ come possibile spiegazione di un esito paradossale, suggerisco che le valutazioni di sé all’interno di un intervallo non clinico, fornite dagli intervistati valutati come disturbati psicologici, possano essere percepite come indicatori `` genuini ‘‘ di un livello apprezzabile di coping e sollievo dai sintomi e non solo come indicazione di un atteggiamento difensivo. Concludo che potrebbe essere più vantaggioso mettere in relazione i risultati ‘paradossali’ con gli aspetti del processo terapeutico, piuttosto che con le disposizioni psicologiche dei clienti.

Ce numéro spécial comprend des études de cas uniques et multiples qui font état du phénomène des résultats « paradoxaux » en psychothérapie se repérant en pratique comme un résultat évaluatif discordant au sein du même cas. Dans un souci de trouver un point d’appui pour mon texte, je me suis efforcé de juxtaposer les cas d’études différentes et ainsi de gagner une vue d’ensemble des contributions et des arguments. À partir de cet exercice, je suggère trois types distincts de résultats paradoxaux empruntant des voies et ayant des implications différentes. En ce qui concerne le concept de « santé mentale illusoire » en tant qu’explication possible du résultat paradoxal, je suggère que les autoévaluations au sein d’un éventail non clinique données par les personnes interrogées estimées psychologiquement perturbées, pourraient être perçues comme un marqueur véritable d’un niveau tolérable de gestion et de soulagement du symptôme et pas seulement l’indicateur d’une attitude défensive. Je conclus qu’il vaudrait peut-être mieux mettre les résultats « paradoxaux » en relation avec des aspects du travail thérapeutique plutôt qu’avec les dispositions psychologiques des clients.

ΠΕΡΊΛΗΨΗ

Το παρόν ειδικό τεύχος περιλαμβάνει μεμονωμένες και πολλαπλές μελέτες περίπτωσης, οι οποίες αναφέρονται στο φαινόμενο των «παράδοξων» αποτελεσμάτων στην ψυχοθεραπεία, τα οποία ορίζονται λειτουργικά ως ασύμφωνες αξιολογήσεις έκβασης μέσα στην ίδια την περίπτωση. Σε μια προσπάθεια να βρεθεί έδαφος για τη δημοσιευμένη απάντησή μου, επιχείρησα να αντιπαραθέσω τις περιπτώσεις από τις διάφορες μελέτες και κάνοντας αυτό να αποκτήσουμε μια επισκόπηση των συνεισφορών και των επιχειρημάτων. Ως αποτέλεσμα αυτής της άσκησης προτείνω τρεις διαφορετικούς τύπους παράδοξου αποτελέσματος με διαφορετικές οδούς και συνέπειες. Συζητώντας την έννοια της «ψευδαίσθησης της ψυχικής υγείας», ως μια πιθανή εξήγηση για το «παράδοξο αποτέλεσμα», προτείνω ότι οι αυτό-αναφορές σε μη-κλινικό φάσμα, που δίνονται από ερωτηθέντες που κατατάσσονται ως ψυχολογικά επιβαρυμένοι θα μπορούσαν να εκληφθούν ως «γνήσιοι» δείκτες ενός αποδεκτού επιπέδου λειτουργικότητας και ανακούφισης συμπτωμάτων και όχι μόνο σαν ένδειξη αμυντικής στάσης. Καταλήγω στο συμπέρασμα ότι θα μπορούσε να είναι πιο συμφέρον το «παράδοξο» των αποτελεσμάτων να σχετίζεται με πτυχές της θεραπευτικής διαδικασίας, και όχι με τις ψυχολογικές διαθέσεις των πελατών.

KEYWORDS:

• Paradoxical outcome
• illusory mental health
• self-rating
• case study

SCHLÜSSELWÖRTER:

• paradoxes Ergebnis
• illusorische psychische Gesundheit
• Selbsteinschätzung
• Fallstudie

PALABRAS CLAVE:

• resultado paradójico
• salud mental ilusoria
• evaluación
• caso práctico

PAROLE CHIAVE:

• esito paradossale
• salute mentale illusoria
• autovalutazione
• studio del caso

MOTS-CLÉS:

• résultats paradoxaux
• santé mentale illusoire
• autoévaluation
• étude des cas

ΛΈΞΕΙΣ-ΚΛΕΙΔΙΆ:

• παράδοξο αποτέλεσμα♦
• ψευδαίσθηση ψυχικής υγείας♦
• αυτό-αναφορά♦
• Μελέτη περίπτωσης

Clinical trials, appraising the efficacy of psychotherapy and counseling, as well as naturalistic studies, evaluating their effectiveness, repeatedly report 60% or more of favorable outcome for clients involved in these treatments (Barkham et al., 2012; Knekt et al., 2016; Wampold & Imel, 2015) . Such judgments are usually made on the basis of self-reports given by client participants on rating forms, designed to measure indications of symptoms of specific mental disorders or of more broadly experienced psychological distress. These measures usually define cut off levels for scores laying in a range considered to show that the individual exhibits an amount of symptoms or troubles implying a need for clinical interventions. Favorable change can then be defined as clinically significant and reliable if the respondent’s scores move from the clinical range (under or over the cut off-line depending on the direction of the scale) and diminish (or increase) with a predetermined number of points on the scale (enough to ensure that the change is not a result of chance) (Jacobson & Truax, 1991). If the cut off-line is not crossed but the change in points is big enough, what you have is reliable change.

Scores on self-report instruments are also increasingly used to monitor the course of ongoing psychological interventions (Lambert et al., 2018; Lutz et al., 2015). When clients repeatedly at regular intervals report on their subjective experience of symptoms or troubles, researchers or practicing clinicians can evaluate whether the progress of the therapy is ‘on track’, that is, whether it follows an expected trajectory as determined by research data from large pools of cases. Usually, those expected courses follow more or less linear or slightly curved progressions, diversions from which are considered as indications that the individual treatment is not ‘on track’, and some corrective actions should be taken.

The logic behind this methodological approach to assess the outcome of psychological interventions, or monitor them, is basically quite straightforward, and the use of self-report instruments a relatively simple and cost-effective alternative to other forms of assessment. Still, questions arise. What if different scales give different indications? What if the respondent, knowingly or not-knowingly, gives misleading points to the items on the scale, that is, points that do not correspond to his or her genuine experience? What if the items are worded in a way that has no meaning to the respondent or a different meaning to him or her than the one intended by the constructors of the scale? What if the respondent’s own experience, as indicated by his or her responses to the items on the scales, does not correspond to observations and impressions of others, be they close relations or professionals involved in treatment or research?

These kinds of questions are the concerns of the present special issue of the European Journal of Psychotherapy and Counselling. The issue includes one introductory text (McLeod, 2021) and four empirical research papers. What the papers at least have in common is that they all employ some kind of a qualitative case study design. Otherwise, they approach the thematic of the special issue from quite diverse angles.

In this published response to the special issue, I focus on the question of possible different interpretations of the meaning of ‘paradoxical’ outcome and make some comments on the concept of ‘illusory mental health’ as an explanation of the phenomenon–the two main themes of the issue. I do this by suggesting a categorization of the cases presented in the different papers on the basis of the type of supposed paradoxical outcome and a comparison of the cases within each category. I offer some ideas on what common features the therapy processes in the cases within the same category might have presented, based on some notions from qualitative process psychotherapy research. There would, certainly, have been much to comment on the individual papers from the perspective of the particular research questions and methodological approaches of each of them but space does not allow for that.

Paradoxical outcome and illusory mental health

There are two main themes in this special issue. One concerns ‘paradoxical’ outcomes in psychotherapy, that is the phenomenon of discordant outcome assessments within the same case. The second theme deals with the concept of ‘illusory mental health’, the notion that people, for different reasons, may hold up a pretense of psychological well-being, while ‘actually’–at least in the eyes of others–not be functioning so well at all.

A paradoxical outcome may manifest itself as a discrepancy between two or more assessment methods, one showing an attainment of treatment goals and another a failure, or an unexpected course of the treatment process, notably when the client moves from the nonclinical to the clinical range of the assessment measure. Instances where the treatment, on an outcome measure, is assessed to be successful but the client is unsatisfied with the process can also be counted for as paradoxical outcomes.

The empirical papers included in this special issue (De Smet et al., 2021; Krivzov et al., 2021; Thoresen et al., 2021; Ward & McLeod, 2021) report on altogether seven (7) cases in which the outcome could be defined as paradoxical (I have counted one from the paper on cases reporting failures in therapy by Krivzov et al.). According to the type of paradoxical outcome, these cases may be grouped as follows (the case J.J. is allotted into two categories according to the phase of her therapy):

Type A: No change on self-rating scale from nonclinical range at intake to nonclinical range at termination but treatment considered as unsuccessful or unsatisfying–Case 5 (Krizov et al.); Rebecca (De Smet et al.); J.J. early treatment phase (Ward & McLeod).

Type B: Change on self-rating scale from clinical range at intake to nonclinical range at termination but treatment considered as unsuccessful or unsatisfying–Paula, Jason, Sophia (De Smet et al.).

Type C: Change on self-rating scale from nonclinical range at intake to clinical range at termination but treatment considered as successful or satisfying–Lisa, on one scale (Thoresen et al.); J.J., case study phase of treatment (Ward & McLeod).

The second main theme of this special issue is the concept of ‘illusory mental health’. The concept is, though, used in different papers in quite various degrees. The concept was originally introduced, defined, and operationalized in a paper by Shedler et al. (1993). Their study design included participants, representing a nonclinical population, who completed the Eysenck Neuroticism scale, a self-report instrument thought to assess a general psychological health-distress factor, and the Early Memory Test, a projective technique aiming at producing a rich material of subjective relational experiences to be used as the basis for a clinical evaluation of the respondent’s mental health or distress. After this, the participants went through a laboratory test in which their cardiovascular reactivity was measured during tasks representing different levels of psychological stress. Using this design, the researchers could identify a group of participants who gave self-reports of mental health but were clinically evaluated as psychologically distressed, and who exhibited high cardiovascular reactivity in stressful situations. This group was in the paper referred to as ‘defensive deniers’ or having ‘illusory mental health’. Of the 41 participants who completed the study, 18 were classified into this category. All in all, 29 (that is 71%!) of the participants, recruited among university students and staff members, were clinically judged as having psychological distress.

Theoretically, the concept of illusory mental health could be connected to the different types of paradoxical outcomes outlined above. In Type A, a client would enter treatment for whatever reason but with a defensive attitude toward manifestations of his or her psychological distress, which would prevail during and after treatment. In Type B the treatment would help the client to cope with ‘surfacing’ expressions of psychological distress but the ‘roots’ of his or her problems would be unsatisfactorily worked on. In Type C, however, the client would enter treatment with a defensive attitude but as a result of the process would become more open to his or her subjective experiences, resulting in an increase in self-rated distress.

Two of the four empirical papers in this special issue (Thoresen et al., 2021; Ward & McLeod, 2021) make more or less articulate use of the concept of illusory mental health when analyzing their cases and the appearance of paradoxical or discrepant outcomes in them. The paper by Krivzov et al. (2021) does not actually deal with the question of paradoxical outcome but describes the lack of failure reports in published psychotherapy case studies–a worthy topic indeed. The paper, though, does identify one clinical case study (Case 5) which could exemplify a paradoxical outcome of Type A, and perhaps the notion of illusory mental health. The paper by De Smet et al. (2021), looking at four cases showing discrepancies between outcome scores and clients’ satisfaction with the therapy, does not use the illusory mental health concept at all when accounting for the observations.

Pathways to paradoxical outcome–a comparison of cases

In the following, I will make a comparison between the reported cases within each of the three types of paradoxical outcome I outlined above.

Type A paradoxical outcome

The three cases in this category entered psychotherapy with self-ratings within the nonclinical range, despite obvious psychological problems of various kinds. Case 5 (Krivzov et al.) is reported to have had a history of dramatic multiple losses of relatives due to health issues, a diagnosis of GAD/health anxiety, and had been treated with heavy psychiatric medication previously. Rebecca (De Smet et al.) entered treatment due to depressive symptoms. J.J. (Ward & McLeod) was said to be at the beginning of therapy a guarded and defended person who suffered high levels of anxiety and feelings of panic, particularly in the context of her relationships with family members.

From the case descriptions, we learn that Rebecca thought her depressive mood had biological reasons, or believed her complaints were caused by her recent change of housing. J.J self-identified as an Adult Child of an Alcoholic (ACOA) and described herself as having a difficult and problematic relationship with her family of origin and, in as much as possible, she had little or no contact with them. I did not find any account of how Case 5 saw the origin of her problems.

During the course of therapy Case 5 is said to have resisted implementing the exposure exercises indicated by treatment guidelines because she did not feel particularly troubled by her anxiety problems. J.J., in the phase of therapy previous to the actual case study, is described as having declined to talk about her early life, or engage in active interventions such as relaxation exercises or art-based activities that might involve being directed by the therapist. No information on how Rebecca responded to or engaged in therapeutic activities during treatment is given.

At termination, Rebecca expressed dissatisfaction with her therapist’s persistence in a perspective on the cause of her problems, which she did not agree with. She had wished for clear help to come to conclusions and see her problems more clearly and thought that the therapist had not challenged her enough. She saw her own changes in life, rather than the therapy, as contributing to the positive change. Yet, seeing the therapist regularly made her feel safe, although the therapy sessions in themselves did not help. In the case of J.J., it is said that despite the positive and supportive relationship that existed between them, the therapist had a strong sense that there was a vulnerable side to the client that she was unwilling to open up. Rather, the weekly sessions became a source of support for J.J. in relation to developing strategies to cope with recurring cycles of conflict with family members. The case description of Case 5 does not include information on the therapeutic relationship, it is, though, mentioned that the client was perceived as having a lack of motivation.

From the information available, it appears that there is much in common between these cases. The clients, to the degree they acknowledged their psychological distress, attributed it to external factors, such as the problematic behavior of close relations, and tended to deal with their difficulties by seeking for control or simply denying their existence. They did not work in therapy actively to achieve change, could decline participating in therapeutic activities, and if they experienced beneficial results, they would rather attribute them to changes in life situations than to the therapeutic process. In spite of this, they seem to have valued the therapeutic relationship as a kind of safe haven.

One might expect that if an analysis of alliance ruptures (Safran & Kraus, 2014) would have been done in these cases, most of them would have had the features of withdrawal ruptures. Such ruptures may often go unobserved by therapists and hence not be dealt with. The main relational problems for the therapists in these cases might have been the difficulty to establish an internal focus for change (Teyber & Teyber, 2016) and how to sensitively stay within the client’s therapeutic zone of proximal development (TZPD) (Leiman, 2012; Leiman & Stiles, 2001), without falling into a collusion (Karlsson, 2004) with the client’s defensive tendencies. The TZPD is the segment of therapeutic development that extends from the client’s current developmental level to the potential one that can be achieved in collaboration with the therapist at a particular moment in therapy (Zonzi et al., 2014). In collusion, the therapist and client would, as an unconscious defensive maneuver, seemingly agree upon working below the level of the TZPD.

Type B paradoxical outcome

The clients from the three cases in this category, Paula, Jason, and Sophia (De Smet et al.), entered treatment exhibiting self-ratings indicating some form of experienced psychological distress, in concordance with professional assessments. At termination, they had moved to the nonclinical range on the self-rating scales but, in spite of this, the clients themselves expressed dissatisfaction with the treatment. Paula entered her therapy with a diagnosis of maladaptive stress reaction with depressed moods, Jason his with depressive complaints. Likewise, Sophia’s presenting problem was her depression.

Paula’s own acknowledged reasons for seeking therapy were her feeling of loneliness and an experiencing that ‘everything felt black’, and the wish to increase security and independence, improve relationships, and be able to handle disappointments. Jason is said to have attributed his problems to a relationship break up due to his own infidelity and sought for a possibility to receive tools to avoid a repetition of such behavior. Sophia described her symptoms as a general sadness that overwhelmed her, and for which she could not find any clear reason.

At termination, Paula’s scores on the global severity index (GSI) showed reliable and clinically significant change, but she, however, expressed dissatisfaction with the treatment and its effects. She was dissatisfied with the therapist, whom she perceived as uncertain and from whom she wished she would have received more advice and answers. The case description does not give any account for the change for better in her self-rating. Jason is reported to have attributed his positive outcome, exemplified by being engaged in a new relationship and experiencing that his work was going well again, more to situational changes than to the therapy process. He would have expected more insight to himself from the therapy, which terminated prematurely. Sophia mentioned several positive changes after therapy but still she was disappointed for not having found a ‘real cause’ for being depressed.

Unlike the Type A category, the clients in the Type B cases, when entering treatment, seemed to have been more aware of their psychological distress and open to face their difficulties. Still, for some reason, the processing of the problems within the therapy was experienced by the clients as unsatisfactory. One might expect that a closer analysis of the therapeutic dialogues would have revealed a pattern where the therapist, for some reason or another, misjudges the scope of the client’s proximal zone of development, thus missing out on opportunities to construct a shared position of observation (Leiman, 2012) from which to explore the client’s difficulties in more depth.

From the perspective of illusory mental health, one would have to conclude that the clients in this category moved, as a result of the therapy, from having been manifestly distressed to an illusory position. A competing, and perhaps more plausible, assumption would be that they actually benefitted from therapy on a level of symptom relief (Stulz & Lutz, 2007) and accurately perceived the missed opportunity to gain a more comprehensive self-understanding.

Type C paradoxical outcome

The two cases included in this category represent instances where a client, at least on one self-rating scale, moved between pre- and post-treatment or subsequent measuring points within the process from a nonclinical to a clinical range. Paradoxically, this change which in a quantitative outcome study would indicate a deterioration in mental health following psychotherapy, was in these cases accompanied by professional assessment of positive change and by an expression of satisfaction with the treatment by clients.

Lisa (Thoresen et al.) had a seven-year history of anorexia nervosa behind her when she entered a new time limited therapy and scored within the nonclinical range on the Eating Disorder Inventory-2 (EDI-2) before therapy but within the clinical range after therapy. The therapist of J.J. (Ward & McLeod), the client with an experience of parental alcohol addiction, entered a training program in Pluralistic therapy, which resulted in a significant shift in her therapeutic approach and a new experience of therapy for J.J.

Lisa had gone through an earlier treatment for her eating disorder which apparently had been beneficial on the level of symptom relief. However, when entering the new therapy, she had told that she has her own ideas about the underlying causes of her illness and expressed that previous therapies have not given her answers to fundamental questions of who she is and why she got sick. From the case description one gets the picture that, in addition to those topics, her uncertainty around self-understanding, particularly around emotional states, and how she was or was not understood by others were areas which she was prepared to work on from the beginning of the therapy.

One significant element of the therapeutic activities was the metaphorical work on Lisa’s two inner voices – a critical voice that would keep her away from life activities to gain control, and a benign voice which was related to her approaching life activities. During the therapy, Lisa and her therapist made use of toy figurines to elaborate and explore the meaning of these metaphors.

By the end of treatment, Lisa herself expressed that she had become more conscious of parts of her earlier experiences and more able to express how she actually felt in different situations. This judgment of a beneficial outcome was corroborated by the therapist and external assessors. In contrast to this, and implying a possible paradoxical outcome, her scores on the EDI-2 was at termination within the clinical range. It should be noted, though, that her scores on the SCL-90-R improved significantly from pre- to post-therapy self-ratings, and that the deterioration of scores on the EDI-2 was mainly (or perhaps only) due to changes on two of the psychological scales (social insecurity and interpersonal distrust) of the instrument.

For J.J., her therapist entering a training program in Pluralistic therapy introduced in some way a new start of her therapy with this therapist. During the earlier three-year phase of the treatment, they had established a trustful relationship, but the focus had been mainly on helping J.J. to cope with her conflictual relationship to her family members. The helpfulness of that phase of therapy was reflected in the finding that J.J.’s self-rated score on the CORE-OM was at the start of the new treatment phase within the nonclinical range.

The new phase of therapy introduced a number of new therapeutic activities. Constructing a timeline diagram, J.J. began to explore significant points in time in her life history, previously not discussed in her therapy but now offering insight to different causal factors of her presenting problems. These included the experience of parental alcohol addiction, childhood emotional and physical abuse, bullying at school, and issues around eating. Another new therapeutic activity introduced by the therapist was an art-based activity during one session. First a short non-guided relaxation exercise was conducted, following which J.J. was invited to draw whatever she wished using art materials made available. No interpretation of the finished artwork was made.

From the point of view of paradoxical outcome, it is noteworthy that J.J.’s scores on the CORE, the 10-item version of which was completed at each other week, showed clinically significant deterioration between the week of the art-based intervention and next measuring point and remained high for the following 5 weeks, before returning to the nonclinical range.

When one looks at the cases of Lisa and J.J. it appears that they have quite a lot in common. Both had a history of treatment which had been beneficial on the level of symptom relief (Stulz & Lutz, 2007) or strengthening of coping skills but unsatisfactory in terms of increased self-understanding (Nilsson et al., 2007). Lisa, when entering a new therapeutic relationship, explicitly worded her wish for such self-understanding as a central goal. For J.J. this was not so obvious, but her therapist had sensed in her a vulnerability which somehow was present though not explicitly addressed. Even more striking in these two cases is the active stance taken by the therapists and the use of metaphorical, non-verbal means in communication and presentation (Levitt et al., 2000).

Concluding remarks

In this special issue, the contributors, using means of qualitative case study methodology, seek to shed light on the phenomenon of ‘paradoxical’ outcomes in psychotherapy, operationalized as discordant outcome assessments within the same case. Some of the papers pay special attention to the concept of ‘illusory mental health’ as a possible explanation for paradoxical outcome. This notion maintains that some people, as a defensive maneuver, hold up a pretense of psychological well-being, while ‘actually’ not functioning all that well. In an attempt to find a ground for a response to the presented findings and views, I endeavored to juxtaposition the cases from the different studies, and by doing that gain an overview of the contributions and arguments.

As a result of that exercise, I suggest three different types of paradoxical outcome. The first category included clients who attributed their problems to external factors and tended to deal with their difficulties by seeking for control or simply denying their existence. The clients in the second category seem to have been more aware of their psychological distress and open to face their difficulties when entering therapy but, still, the processing of the problems was experienced by them as unsatisfactory. In the third category, the clients had in earlier treatment gained symptom relief and/or strengthening of coping skills but no actual self-understanding. In an intensified phase of therapy, conflictual topics were dealt with in more depth which resulted in a temporary or partial increase in self-rated psychological distress.

While accepting the opinion that nonclinical self-reports by people, who in the judgment of others struggle with obvious psychological problems, indicating the absence of an awareness of psychological distress, can be attributed to a defensive attitude, I would suggest that such self-ratings also can be perceived as ‘genuine’ markers of the respondent having attained a tolerable level of coping and symptom relief, and thus as a positive achievement. My sketchy observations on the cases presented in this issue propose that it might be more advantageous to relate ‘paradoxical’ outcomes to aspects of the therapeutic process, rather than to psychological dispositions of the clients. It must be noted, though, that the material provided on process in the case descriptions was so scarce and the information altogether so diverse that a truly reliable comparison of the cases was not possible. To be able to juxtaposition cases from different single case studies we would need a common framework for the information provided. I conclude my response by providing a tentative suggestion for such a framework:

a. formal intake criteria, such as diagnosis, test scores, professional assessment

b. the client’s expressed complaints, concerns and understanding of his/her difficulties

c. the therapist’s initial case formulation and conceptualization

d. the therapist’s theoretical orientation and degree of adherence to some working model

e. detailed enough descriptions of therapeutic activities

f. scores on outcome measures during the therapy process

g. scores on alliance measures during the therapy process

h. multidimensional quantitative and qualitative assessment of outcome on termination and at follow-up

i. semi-structured interviews of therapist and client on the treatment process

As a final note, I recommend that changes in self-rating scores should in case studies not only be inspected on the level of scales but also on the level of individual items. This would make the understanding of changes in the respondent’s experience a lot more detailed and precise.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Jarl Wahlström

Jarl Wahlström, PhD, is Professor Emeritus in Clinical Psychology and Psychotherapy and former director of the Psychotherapy Training and Research Centre at the Department of Psychology, University of Jyväskylä, Finland. His main research interest is in discursive and narrative processes in psychotherapeutic encounters and their relation to the (re)constitution of agency positions and sense of agency as outcome of psychotherapy. On this and related topics he has published in several international journals and books.