Adverse effects of psychological therapy: An exploratory study of practitioners’ experiences from child and adolescent psychiatry

Abstract

The scientific knowledge about adverse effects of psychological therapies and how such effects should be detected is limited. It is possible that children and adolescents are particularly vulnerable and need specific support in order to express adverse effects. In this exploratory study, we used a qualitative approach to explore practitioners’ experiences of this phenomenon. Fourteen practitioners providing psychological therapy within the Child and Adolescent Psychiatric Service were interviewed. Qualitative content analysis was applied to the data. Four overarching categories brought up by the practitioners were identified: vagueness of the concept (reflecting that the concept was novel and hard to define), psychotherapist–client interaction (encompassing aspects of the interaction possibly related to adverse effects), consequences for the young person (including a range of emotional, behavioural and social consequences) and family effects (e.g. professional complications and decreased autonomy for the parent). Professional discussions on these issues could improve psychological therapy for children and adolescents. Based on our findings and previous research, we propose three basic aspects to consider when adverse effects are detected and managed in this context: typology (form, severity and duration), aetiology (hypothesis about the causes) and perspective (adverse effects seen from the points of view of different interested parties).

Keywords

Adverse effects psychotherapy qualitative research psychotherapeutic processes child and adolescent psychiatry

Introduction

Psychological therapy has a powerful potential to accomplish change and is applicable to a range of child and adolescent mental disorders (Chorpita et al., 2011). This type of treatment could potentially also have negative consequences, but adverse effects have – in contrast to beneficial effects – largely been overlooked in the research literature. The scarcity of scientific knowledge leaves the practitioners with the difficult task to detect, manage and prevent such effects. With more knowledge about what types of adverse reactions to expect, practitioners will be better equipped for this task. In addition, a better understanding of adverse reactions to psychological therapy for children and adolescents will enable more adequate information to the young persons and their parents.

Adverse effects have not been systematically reported in psychotherapy trials (Jonsson, Alaie, Parling, & Arnberg, 2014), undermining reliable estimates of their frequency. However, it has been reported that approximately 3–10% of adult patients deteriorate after commencing psychotherapy (Berk & Parker, 2009; Mohr, 1995). Recent online surveys indicate that it is not unusual that clients also have other adverse experiences (Ladwig, Rief, & Nestoriuc, 2014; Leitner et al., 2013; Moritz et al., 2015). A range of adverse effects possibly related to psychological therapies has been reported, concerning multiple domains of the individual’s wellbeing and functioning. These include new symptoms, heightened concern regarding existing symptoms, dependency on the therapist, reluctance to seek future treatment, stigmatisation, a prevailing sick role, disillusion, self-blame and decreased self-efficacy due to not meeting the intervention target (Barlow, 2010; Berk & Parker, 2009; Bystedt, Rozental, Andersson, Boettcher, & Carlbring, 2014; Hadley & Strupp, 1976; Lilienfeld, 2007; Moritz et al., 2015). Also, more extreme forms of adverse experiences related to unethical conduct have been reported, such as sexual harassment or physical violence (Ladwig et al., 2014).

The aetiology of adverse effects of psychological therapies is likely to be multifactorial. First, the techniques used might cause adverse reactions. Available data suggest that some psychological therapies can do more harm than good (e.g. critical stress debriefing, recovered memory techniques and boot camp interventions) (Lilienfeld, 2007). Other therapies may produce transient states of distress, such as increase in symptom severity during exposure therapy (Olatunji, Deacon, & Abramowitz, 2009). Second, the treatment format might be linked to adverse effects. Long-term psychotherapy increases the risk of a dependency that may affect self-mastery (Berk & Parker, 2009), while group therapy attendance may become such an important part of an individual’s life that it leads to a perseverance of symptoms (Nutt & Sharpe, 2008). Third, factors related to the patients should be considered. It has, for instance, been suggested that a dependent personality style, limited social support and chronic and disabling conditions increase the risk of excessive dependence and maintenance of a sick role (Berk & Parker, 2009). Fourth, characteristics of the therapist can be of relevance. Kraus, Castonguay, Boswell, Nordberg, and Hayes (2011) found that the average patient deteriorated significantly for between 3% and 16% of therapists, depending on the symptoms treated. Finally, features of the service setting – including organisation and safety culture – can be of significance.

Apart from the complex aetiology, there are a number of other challenges related to detection of adverse effects in clinical practice. It is as yet not clear which adverse effects to look for, when they can be expected to emerge, how patient should be prompted to report such effects and to what extent other individuals than the patient are liable to be adversely affected. In addition, the patient, the therapist, significant others and the society can have different perspectives and expectations on the outcome (Strupp & Hadley, 1977). Thus, there is a need for a clearly defined terminology and useful models to describe these phenomena. Hoffmann, Rudolf, and Strauß (2008) distinguished between side effects of an adequate therapy, side effects related to unprofessional practice, side effects related to mismatch between the patient’s and the therapist’s personalities, and harm caused by unethical conduct of the therapist. Recently, Linden (2013) and Linden and Schermuly-Haupt (2014) developed a theoretical framework for unwanted events, including definitions of side effects and malpractice effects. Unwanted events are defined as all negative events that occur parallel to or in the context of treatment. Unwanted events caused by the treatment are defined as adverse treatment reactions. An adverse treatment reaction is categorised as a side effect when the treatment is correctly applied and a malpractice reaction when the treatment is not correctly applied. Side effects occurring routinely constitute the ‘side effect profile’ of that treatment, which the patients have the right to be informed about (Linden & Schermuly-Haupt, 2014).

Previous research on adverse effects of psychological therapies has mainly focused on adults. There are several reasons why it is important to also study this phenomenon specifically in children and adolescents. First, the rate of deterioration during treatment might be even higher in children and adolescents than in adults. A study investigating psychotherapy change trajectories and outcome for children and adolescents reported that 14% of cases in managed care settings and 24% of cases in community mental health settings deteriorated (Warren, Nelson, Mondragon, Baldwin, & Burlingame, 2010). A recent study of psychodynamic psychotherapy for adolescents and young adults found that while a majority improved, 9% deteriorated (Nemirovski Edlund & Carlberg, 2014). Second, a lack of influence and power during these formative years could add further to the negative impact of adverse effects. Children may have difficulty in expressing symptoms and effects of the treatment, which may lead to under-reporting. While adults are free to discontinue treatment if deterioration occurs or in case of no treatment effect, children could depend on the judgements made by the therapist and the parents. Third, the impacts of the adverse effects may be greater as the individual is likely to be particularly vulnerable during the rapid cognitive and emotional development during childhood and adolescence. Fourth, in the treatment of children and adolescents, certain ethical issues arise (Koelch & Fegert, 2010). The treatment involves not only the child but often requires the involvement of the family. Thus, the integrity and autonomy both of children and other family members can be affected. In addition, there are aspects of confidentiality that may conflict with the protection of the child, that is, in the context of child abuse or if the child has asked the therapist not to share certain information with the parents (Koelch & Fegert, 2010; Tan & Fegert, 2004).

In this study, we used a qualitative approach to explore practitioners’ experiences of the possible adverse effects of psychological therapy with children and adolescents. We choose the practitioners’ perspective at this explorative stage in order to increase the likelihood to detect signals of adverse effects, by taking advantage of the practitioners’ large number of encounters with patients. In addition, asking children and adolescents about negative experiences related to their treatment might involve unknown risks. In order to be ethically justified, we deem that such a study requires a more thorough prior understanding of the phenomenon.

Method

Participants and procedure

All 12 clinics of the outpatient Child and Adolescent Psychiatric Services in Uppsala, a middle-sized county council in Sweden, were informed about the study by the authors. Each clinic was requested to provide participants for at least one interview. Interviewees should be practitioners who had undergone psychotherapeutic training, worked actively with psychotherapy when invited and had at least 2 years of experience conducting psychotherapy. This recruitment strategy was adopted in order to include a diversity of experiences from a range of clinical settings.

In all, 14 practitioners (mean age, 49 years; 12 women; nine psychologists/three social workers/two nurses/; nine mainly from cognitive behaviour therapy and five mainly from psychodynamic psychotherapy) from seven clinics were included. Their mean experience of providing psychological therapy was 19.5 (standard deviation (SD) = 14.1) years. One clinic did not provide psychological therapy, and four clinics did not reply to the request. Due to personal preferences of the interviewees and/or for practical reasons, five interviews were performed with single individuals and three as focus groups with two to four members from the same unit, using the same interview guide and lasting for 45–60 minutes. The interviews were performed and audio-recorded at the clinical units with one main interviewer and one co-interviewer (last year students in clinical psychology who changed roles between interviews). The role of the co-interviewer was to follow the discussion, to ask supplemental follow-up questions and to take notes. This strategy was adopted in order to increase the quality of the data and to minimise the risk of misconceptions.

Interview and data analysis

An interview guide was created and pilot-tested. Overarching questions were selected that would provide a framework for discussion, while introducing a minimum of information about potential adverse effects. At the start of the interview, the therapists were informed of our definition of adverse effects: ‘non-intentional effects perceived as negative, uncomfortable or harmful by patient, therapist or family’. The definition was inspired by the definition of side effects by Hoffmann and colleagues (2008) and the concept of adverse treatment reactions (Linden, 2013; Linden & Schermuly-Haupt, 2014), in combination with the outline of different perspectives on treatment outcome proposed by Strupp and Hadley (1977). The definition and the interview guide were designed to be non-directive and include also less severe distress that might not spontaneously be regarded as an adverse effect by all therapists. The first question concerned experiences of adverse effects of psychological therapy. Following the interview guide, questions were then asked (in adapted order and phrasing) on types of unwanted effects of psychological therapy (experienced or reported by patients or colleagues from previous treatments), type of psychological therapy, frequency, age and diagnosis/problems, severity, impact on the patient, own handling of the adverse effect, hypothesised causes and adverse effects experienced by others than the patient.

Recruitment, data collection, transcription and analysis were carried out in parallel during 8 weeks from February to April 2013. Verbatim transcription of each interview was performed by the main interviewer, and the document was then scrutinised by the co-interviewer and adjusted after consensus. All transcriptions were anonymized, and each participant and unit was assigned an arbitrary number. The transcription resulted in 82 pages of text in total. Qualitative content analysis was adopted (Elo & Kyngas, 2008; Graneheim & Lundman, 2004). The choice of method was guided by the explorative nature of this research, as it is not yet clear what type of adverse effects that might be observed by practitioners.

During this process, transcriptions were read several times, whereby relevant and meaningful units were gradually identified (254 units were finally aggregated) and condensed. The first interview was coded and re-coded by both interviewers in order to ensure that the coding was done in a similar way by both interviewers, while the following interviews were coded by either of them. The codes were then formulated into subcategories. In the final step, overarching categories were created out of the subcategories. Two senior authors supervised and critically discussed the coding and categorising process and formulated alternative ways until consensus was reached. Two examples of the categorisation procedure are presented in Table 1.

Table 1.

Illustration of the analytic procedure from meaningful units to overarching categories.

Meaningful unit	Aggregation/ condensation	Code	Subcategory	Overarching category
I think that one unwanted effect that I see . . . it is that we sometimes take over the parental role . . . the parents can go home while the kids are here or, yes, that we sometimes do the parental tasks.	The responsibility is transferred from parents to therapist	Parental lack of responsibility	Parent less autonomous	Family effects
Well it turns into a negative spiral for the whole family system. The parents may feel that they have failed. They feel ashamed and guilty of not being able to get the child to school. ‘What have I done that was wrong?’	Parents react by feeling ashamed and guilty and with a sense of failure	Parental emotional reactions	Family members affected	Family effects

Ethical approval

All participants gave their written informed consent to take part in the study. In order to protect the identity of the participants, each participant was assigned an arbitrary number. The confidentiality of information provided was assured. The study was approved by the Regional Ethical Review Board of Uppsala, no. 2012/486.

Results

The following presentation sets out from the categories and subcategories that were brought up by the qualitative content analysis, summarised in Table 2. The categories ‘Vagueness of the concept’ and ‘Psychotherapist–client interaction’ reflect themes related to the concept of adverse effects, while the categories ‘Consequences for the child’ and ‘Family effects’ represent specific adverse effects brought up by the practitioners. Frequencies are described by the words ‘one’, ‘some’ or ‘many’. ‘Some’ indicates at least two but less than the majority of the interviewees, while ‘many’ denotes the majority of the interviewed practitioners. The presented quotes were chosen on the basis that they reflected the categories and subcategories.

Table 2.

Categories and subcategories.

Category	Subcategory
Vagueness of the concept	New way of thinking
	What does adverse effect mean?
	Who determines when an adverse effect has evolved?
	Who is affected?
	Expected or not?
Psychotherapist–client interaction	Premature termination or delayed treatment
	Suboptimal alliance
	Lack of compliance
Consequences for the child	Increasing or emerging symptoms
	Distress and maladaptive reactions
	Social consequences
	Child less autonomous
Family effects	Family members affected
	Parental professional complications
	Parent less autonomous

Vagueness of the concept

New way of thinking

When presented to the concept of adverse effects, many interviewees at first found it unfamiliar, especially since their usual approach was to try to avoid complications and to solve problems. Thus, some thought that they might have failed to observe adverse effects. During the first part of the research interviews, it was commonly regarded as a rare phenomenon:

This is not something that I have been reflecting upon that explicitly, or using such words. But it is something that you discuss during supervision and with colleagues. It is there, but one does not have a definition, not like a concept. (Participant 12)

What does adverse effect mean?

Since the concept implied a new way of thinking, defining the concept was found problematic. Many interviewees found it difficult to separate adverse effects from symptoms and reactions that should rather be attributed to the underlying mental disorder. For example, one interviewee mentioned that a tendency to withdraw from social contact could be due to either aggravation of symptoms of the mental disorder or a consequence of a treatment that requires absence from school and other activities during leisure time.

Who determines when an adverse effect has evolved?

The client and the psychotherapist may have different opinions about whether symptoms, behaviours and emotions should be defined as adverse effects or not. Moreover, some therapists had the experience that clients may perceive a certain phenomenon differently during different phases of the treatment. What was regarded as a strain during the psychotherapy could afterwards be regarded as ‘well worth the pain’. The ambiguity of the concept was illustrated by the example that adolescents may perceive parental restrictions during family therapy as painful and possibly as adverse effects, but simultaneously – or after the therapy – they may perceive and even express wishes for such distinct guidance. Furthermore, newly acquired skills as a result of the treatment, such as self-assertive behaviour, could be viewed negatively by staff at school but positively by the child.

Who is affected?

Some therapists pointed out that the adverse effects may not only afflict the identified patient but also the parents and the siblings. For instance, a child’s growing autonomy may be perceived negatively by other family members. There was uncertainty whether the latter should be regarded as an adverse effect:

If the problem does not have its roots in the patient but in the family system, then there may be desirable changes here but unwanted effects for the family. (Participant 5)

Expected or not?

An increased level of anxiety was regarded as an expected and even necessary effect of exposure-interventions, which may even confirm that the treatment is effective:

I think that this is the key to success – so to speak – but obviously there are parents who find it difficult. (Participant 8)

It was described that the increased anxiety is characteristic of the initial phase of the treatment and that it usually is of short duration:

It is often much more difficult at the start and it is just the kind of unwanted treatment effect that may appear during the exposure phase. (Participant 2)

Psychotherapist–client interaction

At all interviews, there were comments about the importance of the psychotherapeutic cooperation and alliance between the involved family members and the therapist. Shortcomings in this respect were thought of as adverse effects or causes or consequences of adverse effects.

Premature termination or delayed treatment

Many interviewees regarded a prematurely interrupted treatment – or a treatment that for some reason did not even start – as unwanted. Introduction of delicate information and a formal report to the child protection services were mentioned as two reasons for a premature termination. A discrepancy between therapist and the family regarding the set-up of the treatment was mentioned as another reason. Further reasons could be that the patient did not find the contact with the therapist satisfactory or that the treatment was not effective:

Well, maybe that you do not come back. Because you think that there is no point. I have a girl, for instance, who just came for one or two sessions and she thought that nothing happened. (Participant 1)

Some therapists pointed out that the personal reason for a premature termination is not always obvious, and it may even be difficult to find out whether it should be categorised as an adverse effect or not. The patient may simply feel well and not motivated to continue.

Suboptimal alliance

Some therapists told about patients who felt uncomfortable with them. The patient may have felt misunderstood or treated in a bad way. In other cases, no rapport had been established or the patient had simply preferred a therapist of the same gender. Many therapists had experienced conflicts with patients. Different reasons were mentioned: the patient did not want to involve the parents; the patient had the impression that the therapist gave priority to the parental perspective; the therapist had – due to high work load – mixed up information about the patient with information about another person; the therapist put too much pressure on the patient. The therapist could be aware of the conflicts through a negative counter-transference or as a subtle feeling of discomfort.

Lack of compliance

Some therapists reported about patients who had not understood why they were to start psychotherapy, what psychotherapy implies or what they were expected to do as a patient. Such circumstances seemed to have influenced both motivation and treatment effects. Parent-related issues were mentioned by many as factors that might complicate the treatment procedure: parents may have found it difficult to understand how their own behaviour could contribute to maintain or increase the problems of the patient; parents may have lacked insight about the effects of adversities at home; parents may have reacted negatively at confrontations involving own shortcomings:

. . . parents may find it very difficult to understand that their way of caring for their child could worsen the problems. And when you confront them with adversities, social adversities that is, complications will almost be expected. (Participant 8)

Consequences for the child

Increasing and emerging symptoms

Either increase in the target symptoms or the emergence of new symptoms, expected or unexpected, was described. The most common description was increased symptoms, like self-harm and anxiety, often related to exposures or to the termination phase:

I work with those who are self-destructive and this behaviour may re-occur at exposures and it is actually a rather serious consequence. (Participant 2)

Children with social phobia or obsessive-compulsive disorder sometimes feel terrible during exposure. I have had patients who vomit due to their distress and who experience a strong increase of their anxiety. (Participant 9)

Distress and maladaptive reactions

A range of emotional reactions during treatment was reported: the patient may feel betrayed by the parents because they did not take action earlier, react against psychotherapeutic demands or feel disappointed when the effect of the treatment does not appear as quickly as expected:

They may be frustrated when exposed to new demands about participation and maybe also to perform tasks between sessions in order to get better. (Participant 2)

New unexpected problems may also arise that could be related to the treatment:

A teenager who has always been calm at school may suddenly mess around. (Participant 10)

Some therapists reported that new insight, for instance, about parental shortcomings may evoke distress:

An adolescent has low self-esteem and thinks that there is something wrong with them . . . Because they have had these signals at home. Nagging and complaining and downgrading. And then they start to grow and see that Mummy maybe isn’t fair, or Daddy or whatever. (Participant 10)

Social consequences

The treatment might also change or interact with the social context in a way that could lead to adverse outcomes. One interviewee reported that the patient might lose contact with school and/or other important social networks like peers (mentioned by many) for a while. This may be a direct result of a treatment that requires absence from school, due to the severity of the illness, to enable the child to follow the treatment guidelines (one example) or an indirect effect related to aggravation of the symptoms during psychotherapy (many examples). Some interviewees mentioned depression as a condition where the social network is a crucial issue. Home assignments during psychotherapy involving taking up contacts with friends may fail, which may in turn decrease the motivation for psychotherapy:

Depression to a high degree concerns renewal of social contacts and such things and a negative consequence may be that friends are not willing at that time to take up the contact again. (Participant 2)

Child less autonomous

Some reported that patients may start to rely exclusively on their therapists instead of parents or friends:

I had a very distinct comment from a patient that ‘I do not need any friends because I talk to you’. Then you have really done somebody a disservice. (Participant 13)

Some therapists described the children’s need for confirmation from the therapist about how they feel and how everything is developing:

At times they may be lulled into an attitude of: ‘How are you going to solve this problem for me?’ (Participant 2)

Termination was described as a particularly sensitive period for this kind of phenomena. Many therapists reported that negative reactions – like disappointment and increased anxiety, distress and sadness – coincided with the termination phase or even with suggestions about termination. A feeling of being abandoned could also be evoked when a therapist cancelled a session due to sickness. Such reactions could appear quite unexpectedly:

During termination everything may appear. That is, it may turn out to be very tough to leave, perhaps unexpectedly. Everything about separation became such a huge issue, which might not have been apparent before. (Participant 14)

Family effects

There were reports about adverse effects not only for the child but also for parents and siblings and even for other parts of the social network. Such effects could emerge even for parties not directly involved in the treatment.

Family members affected

Many therapists found that the consequences for other family members differed depending on the age of the patient; if the patient was young, the treatment to a higher degree involved others:

I think that the family and the parents get more negative consequences if the child is young, because in such cases they are much more involved. (Participant 2)

This can also mean that siblings get less attention and have to manage on their own. One therapist said that some treatments imply that the parents together with their child approach situations that the child finds uncomfortable and that the parents might perceive that they are hurting the child when doing so:

It can be rather unpleasant for a parent to think that you hurt your child in some way. (Participant 4)

Some therapists reported that conflicts between the child and the parents might emerge in relation to the homework that is a part of the treatment. Feelings of guilt and shame or inadequacy may also arise, at times enhanced by a notion that the problems of the child are worse than they really are since the child receives psychiatric treatment.

Many therapists reported parental reactions of anger in different situations. This may be prompted by an apprehension that the focus of the therapy is wrong or that delicate matters are brought up. These adverse effects can have both short- and long-term consequences since they are not only perceived as unpleasant but can also negatively influence the treatment.

Parental professional complications

One therapist told about the parental problem of being absent from work in order to attend the psychotherapeutic sessions. Some treatments imply a parent being absent from work for an extensive period of time to be able to care for their child and one therapist told about parents who as a consequence had lost their employment or been forced into severe financial problems.

Parent less autonomous

Many therapists had the impression that they had been ascribed a substantial part of the parental role. As a consequence, there were no changes at home and the child lost his or her trust in the parents:

We want to achieve a re-establishment of parental authority and an adaptive situation over time. If we are the authorities there will be no change at home and everything will work out as before. There will be no change in the family. And the child will get the impression that Mom and Dad can’t handle the situation. (Participant 11)

Many therapists told about parental feelings of incompetence, especially when the therapist failed to convey a sense of progress or failed to empower the parents:

If we fail to convey what is well done . . . that the patient and the parents may feel that this is hopeless, it doesn’t get any better, we feel that we are bad parents or the child does not perform good enough. Because . . . yes, they don’t feel empowered but helpless. (Participant 11)

Discussion

In this study, we explored practitioners’ experiences of adverse effects of psychological interventions in child and adolescent psychiatry. Some of the themes might be applicable independent of age, meaning that they are relevant for both children and adults. For instance, increasing and emerging symptoms corroborate findings from recent studies on adults (Moritz et al., 2015; Rozental, Boettcher, Andersson, Schmidt, & Carlbring, 2015). Dependence/loss of autonomy, deterioration and impact on other life domains were also found in an online survey among cognitive behaviour therapists (Bystedt et al., 2014), although the age groups treated by the therapists taking part in the survey were not specified.

Other aspects seem to be more specific to the child psychotherapy perspective, in particular issues related to the family. Thus, adverse reactions may refer to any family member, most often exemplified as the parents. Their child – and the therapist – may confront the parents with difficult and painful matters. The parents may also feel guilty and ashamed when exposed to such confrontations. Issues about autonomy also have a special significance in relation to childhood. The risk of becoming too dependent upon the therapist may be enhanced by the age and maturity inequality. Moreover, the child – and/or the parents – may perceive a conflict between the parental and the therapist’s authority. Demanding therapies may imply that the child, at least during a certain period, will lose contact with the social network, like peers. Behavioural changes as a result of personal growth during a successful therapy may be perceived as adverse by parents or teachers.

Aspects of the therapist–client interaction were explicitly mentioned as possible causes or consequences of adverse effects, including early termination and poor alliance and compliance. These results are in line with the category of side effects related to mismatch between the patient and the therapist proposed by Hoffmann et al. (2008). In a recent online questionnaire on patient’s perspectives on psychotherapy, 22.3% perceived their therapeutic relationship as poor (Leitner et al., 2013). In addition, 31.8% of the respondents who had terminated their treatment had done so prematurely. These data suggest that poor quality of the therapist–client relationship is a frequently occurring problem. The involvement of significant others in treatment of children can obviously make the interaction even more complicated.

Although many practitioners found our concept of adverse effects vague, they provided ample and diverse examples on the topic when prompted. This is consistent with previous research regarding the difficulty of detecting adverse effects (Dimidjian & Hollon, 2010; Hatfield, McCullough, Frantz, & Krieger, 2010). Several components contributed to their initial doubtfulness. First, the practitioners had not reflected on such occurrences in terms of adverse effects; thus, it was a new way of thinking. Second, they found it difficult to determine whether an emotional, behavioural and social reaction should be regarded as an adverse effect or as a manifestation of the illness itself. Third, they pointed out that patients, therapists and caregivers could view adverse effects differently. This is in line with the different perspectives outlined by Strupp and Hadley (1977) and raises the question of how different perspectives should be taken into consideration when adverse effects are assessed. The practitioners pointed out that certain effects were expected and even served as a proof that the therapy was efficient. Still, such experiences could be perceived as adverse by the patients.

The difficulties in identifying and recognising adverse effects emphasise the importance of a greater awareness among therapists on these issues. Such improved awareness, better definitions of the concept of adverse effects and a distinct framework for describing these phenomena may contribute to further improvement of psychological therapy for children and adolescents. In all humility, we would like to propose an outline to a meta-model of basic aspects that could assist practitioners in detecting adverse effects in children and adolescents. We acknowledge recent work, which has provided frameworks for definitions and tools for monitoring of adverse effects. We take our starting point in previous work, trying to integrate our own experiences from this study. At the same time, we apply our proposed model as a structure for the upcoming discussion.

We suggest three aspects as tentative basic elements of such a model: typology, aetiology and perspective. Typology concerns the types of adverse effects in focus, the degree of severity, the timing and the duration. This information is obviously indispensable when monitoring adverse effects: the practitioner must know what to look for, when to look for it and how critical it might be. Several different types of adverse effects were reported by the practitioners in this study, including increase in ongoing symptoms, appearance of new symptoms, loss of autonomy and social consequences.

Aetiology refers to any casual factors, including the treatment used and aspects of the alliance and interaction between patient (and/or other family members) and the therapist. Also, the service setting or other contextual factors may be of decisive influence; unsolved conflicts between social actors (e.g. school, child protection services and child psychiatry) may obstruct the process and create a psychological substrate promoting adverse effects. Likewise, parental conflicts may influence the outcome – and adverse effects – of psychological therapy. The construct of aetiology also refers to temporal aspects like unplanned interruptions and termination. While scientific identification of mechanisms of harm requires experimental manipulation of the variable in question (Dimidjian & Hollon, 2010), the chance of preventing adverse effects in forthcoming patients can improve if practitioners actively try to identify the sources of the problems.

It is important to note that some adverse effects may be expected or even unavoidable. Other might be avoided by improved safety culture, supervision or changes in service structure and organisations. From a clinical perspective, it could be particularly important to distinguish between avoidable reactions and reactions that are an expected and unavoidable result of the treatment (e.g. increased anxiety during exposure). This distinction is akin to the distinction between side effects and malpractice effects proposed by Linden and Schermuly-Haupt (2014) and the distinction between side effects of an adequate therapy and side effects related to unprofessional practice or harm related to unethical conduct proposed by Hoffmann et al. (2008). However, it should also be noted that some risks related to correctly applied treatment might be avoided, if the treatment can be modified for this purpose.

Perspective refers to the person defining when an adverse effect has evolved, the theories applied to describe it and any social structures involved in these processes, and the individual affected. As pointed out by Strupp and Hadley (1977), the professional, the client and the society have different perspectives on treatment outcome. This study obviously emanates from the therapist perspective. This is illustrated by examples where therapists describe behaviours or emotional reactions as ‘expected’, given the assumptions inherent in the psychotherapeutic theory applied. The connotations related to the word ‘expected’ means that it is doubtful from a therapist perspective whether a certain reaction should be regarded as an adverse effect or not, whereas a patient may clearly have categorised the same reaction as such. As indicated in the vignettes, such a categorisation of the patient may be withdrawn over time, highlighting the importance of time for conceptualisation. We believe that it is important to be attentive also to less severe and transient distress, in order to decrease the risk of premature termination.

Structured around the three basic aspects of our model, we briefly outline some ideas for further clinical discussion. Monitoring systems can alert practitioners to potential adverse effects and might improve outcome (Lambert, 2007). Linden (2013) has provided a checklist for the assessment of unwanted events and adverse treatment reactions, taking context, relatedness to treatment and severity into consideration. Other extensive measures of side effects of psychotherapy have also been developed recently (Ladwig et al., 2014; Parker, Fletcher, Berk, & Paterson, 2013). However, available measures are mainly developed for adults. The nature of adverse effects may differ in the treatment of children and adolescents, and therefore, a developmentally sensitive checklist especially suited for this age group could be required. In addition, measures of adverse effects need to be tailored for specific interventions and patient groups.

Given the relatively high rates of deterioration reported in children and adolescents treated with psychological therapies (Nemirovski Edlund & Carlberg, 2014; Warren et al., 2010), it is also critical that therapists treating young people are attentive to deterioration. If the severity of the condition is regularly measured on a continuous scale during treatment, deterioration might be detected by adopting a reliable change index (Jacobson & Truax, 1991). For instance, Warren et al. (2010) have developed predicted change trajectories and a warning system to identify cases at risk of treatment failure among children and adolescents.

Many interviewees described difficulties in determining whether a symptom is an adverse effect or a manifestation of the underlying disorder, which complicates detection. As pointed out by Dimidjian and Hollon (2010), harmful effects can be masked and therefore difficult to detect in disorders that either follow a deteriorating course or improve over time. However, we concur with Linden and Schermuly-Haupt (2014) that the practitioner should at least try to rate the probability that the adverse effect is related to the treatment (e.g. unrelated, probably unrelated, possibly related, probably related, definitely related).

In treatment of children and adolescents, the patient and the parents might observe and report different problems. These different perspectives could be addressed using different categories of informants. Children – due to lack of influence and power – could benefit from external support concerning the identification and naming of adverse effects. The therapist – and child psychiatry in general – has a special responsibility for contributing to formulating the child perspective on these issues. Hypothetically, the prevailing theoretical perspectives and apprehensions of a certain phenomenon may be decisive for defining adverse effects. Professional attitudes – based on psychotherapeutic theories or even prevailing prejudices – may contribute to a resistance against accepting patient reports. Such hinders to detection of adverse effects might be amplified if the patient feels that the alliance with the therapist could be undermined by negative feedback. Systems enabling patients to report complaints anonymously, or to independent instances, might overcome such barriers. In addition, it is important that knowledge and skills regarding patient safety are continuously maintained at a high level within child and adolescent psychiatric settings.

Limitations

It should be emphasised that this is an exploratory study aiming at finding themes of relevance for professionals in relation to adverse effects of psychological intervention during childhood. The results presented here should be viewed in the light of the general difficulties of studying and defining this topic.

We report interviews from a fairly small number of practitioners. The average experience of conducting psychological therapy was almost 20 years, illustrating that they had experienced a large number of encounters with patients. However, given the host of psychological therapies, age groups and specific problems presented within child and adolescent psychiatric care, it is unlikely that we have detected all aspects of importance. Potential adverse effects related to the treatment might also take different forms depending on the patient’s social context, which is likely to vary considerably. Other scientific approaches are necessary for drawing conclusions about adverse effects in relation to specific disorders, specific treatments or specific age groups and for evaluating the severity or frequency of various adverse effects.

It should be noted that four of the clinics did not respond to the invitation to participate, which to some extent could have influenced the results. On a related note, it is likely that the practitioners volunteering for this study were more aware of this phenomenon than the average therapist.

Another limitation – or rather choice of perspective – is that only practitioners were interviewed. This is mainly a choice made from ethical considerations. Reports on side effects of psychotherapy during childhood are scarce, and we deemed it necessary to first get an overview using professional experiences. It is obviously an ethical challenge to approach children in psychiatric care, and researchers need to be well prepared in order to be able to formulate issues of relevance, at the same time avoiding undue intrusions.

Finally, it is possible that our own perspective to some extent influenced the results. Our aim was to explore practitioners’ experiences of adverse effects of psychological therapy. To safeguard against over interpretation, all four authors discussed the coding and categorising process and formulated alternative ways until consensus was reached.

Conclusion

The concept of adverse effects related to psychological interventions seems to be unfamiliar and vague to many practitioners. Although practitioners are committed to high ethical standards, there is a risk that adverse effects or deterioration sometimes go unnoticed. Professional discussions among therapists on these issues have the potential to improve psychological therapy for children and adolescents further. Developing and implementing concepts that cover different aspects of these phenomena could facilitate such discussions. The findings from our exploratory study suggest that not only the patient but also close relatives may be affected negatively by psychological therapy for children and adolescents. Consequently, practitioners should take into consideration all parties that may be influenced. Future research should explore adverse effects from the perspective of the patients and their families. In addition, it is important to study the typology and aetiology of potential adverse effects of psychological interventions for specific interventions, patient groups and age groups.

Footnotes

Acknowledgements

We thank the participating practitioners for generously sharing their experiences with us. UJ had the original idea and supervised the design and execution of the study. JJ and EN collected and analysed the data. All authors contributed to the conception, design, interpretation of the data, drafting the manuscript and critical revision of the text. All authors have approved of the final version of the manuscript.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Author biographies

Ulf Jonsson, PhD, is a licensed psychologist and an Associate Professor of Child and Adolescent Psychiatric Research at Uppsala University. His research has mainly focused on the transition from adolescence to adulthood for people with mental disorders. He is currently employed as a researcher with the National Board of Health and Welfare, Sweden.

Josefin Johanson, MSc, is educated in clinical psychology and is currently working as a school psychologist. Her clinical interests include school environment and mental health, consultation, psychological assessment and organizational psychology.

Elin Nilsson, MSc, is educated in clinical psychology and is currently engaged in the work conducted by the National Board of Health and Welfare, Sweden, regarding safety issues related to psychological treatment.

Frank Lindblad, MD, PhD, is a Professor Emeritus of Child and Adolescent Psychiatry at Uppsala University. His current research follows three main lines: stress research, mainly involving psycho-physiological markers; epidemiological research focusing on developmental vulnerability; and evaluation of allegations concerning child sexual abuse.

References

Barlow

D. H.

(2010). Negative effects from psychological treatments: A perspective. The American Psychologist, 65, 13–20. doi:10.1037/a0015643

Berk

Parker

(2009). The elephant on the couch: Side-effects of psychotherapy. The Australian and New Zealand Journal of Psychiatry, 43, 787–794. doi:10.1080/00048670903107559

Bystedt

Rozental

Andersson

Boettcher

Carlbring

(2014). Clinicians’ perspectives on negative effects of psychological treatments. Cognitive Behaviour Therapy, 43, 319–331.

Chorpita

B. F.

Daleiden

Ebesutani

Young

Becker

K. D.

Nakamura

B. J.

. . . Starace

(2011). Evidence-based treatments for children and adolescents: An updated review of indicators of efficacy and effectiveness. Clinical Psychology: Science and Practice, 18, 154–172.

Dimidjian

Hollon

S. D.

(2010). How would we know if psychotherapy were harmful? The American Psychologist, 65, 21–33.

Elo

Kyngas

(2008). The qualitative content analysis process. Journal of Advanced Nursing, 62, 107–115. doi:10.1111/j.1365-2648.2007.04569.x

Graneheim

U. H.

Lundman

(2004). Qualitative content analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurse Education Today, 24, 105–112. doi:10.1016/j.nedt.2003.10.001

Hadley

S. W.

Strupp

H. H.

(1976). Contemporary views of negative effects in psychotherapy: An integrated account. Archives of General Psychiatry, 33, 1291–1302.

Hatfield

McCullough

Frantz

S. H. B.

Krieger

(2010). Do we know when our clients get worse? An investigation of therapists’ ability to detect negative client change. Clinical Psychology and Psychotherapy, 17, 25–32.

10.

Hoffmann

S. O.

Rudolf

Strauß

(2008). Unerwünschte und schädliche Wirkungen von Psychotherapie: Eine Übersicht mit dem Entwurf eines eigenen Modells [Unwanted and damaging effects of psychotherapy. An overview with a draft of a new model]. Psychotherapeut, 53, 4–16.

11.

Jacobson

N. S.

Truax

(1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12–19.

12.

Jonsson

Alaie

Parling

Arnberg

F. K.

(2014). Reporting of harms in randomized controlled trials of psychological interventions for mental and behavioral disorders: A review of current practice. Contemporary Clinical Trials, 38, 1–8. doi:10.1016/j.cct.2014.02.005

13.

Koelch

Fegert

J. M.

(2010). Ethics in child and adolescent psychiatric care: An international perspective. International Review of Psychiatry, 22, 258–266. doi:10.3109/09540261.2010.485979

14.

Kraus

D. R.

Castonguay

Boswell

J. F.

Nordberg

S. S.

Hayes

J. A.

(2011). Therapist effectiveness: Implications for accountability and patient care. Psychotherapy Research, 21, 267–276.

15.

Ladwig

Rief

Nestoriuc

(2014). What are the risks and side effects to psychotherapy? – Development of an Inventory for the assessment of Negative Effects of Psychotherapy (INEP). Verhaltenstherapie, 24, 252–263.

16.

Lambert

(2007). What we have learned from a decade of research aimed at improving psychotherapy outcome in routine care. Psychotherapy Research, 17, 1–14.

17.

Leitner

Märtens

Koschier

Gerlich

Liegl

Hinterwallner

Schnyder

(2013). Patients’ perceptions of risky developments during psychotherapy. Journal of Contemporary Psychotherapy, 43, 95–105.

18.

Lilienfeld

S. O.

(2007). Psychological treatments that cause harm. Perspectives of Psychological Sciences, 2, 53–70.

19.

Linden

(2013). How to define, find and classify side effects in psychotherapy: From unwanted events to adverse treatment reactions. Clinical Psychology and Psychotherapy, 20, 286–296. doi:10.1002/cpp.1765

20.

Linden

Schermuly-Haupt

M. L.

(2014). Definition, assessment and rate of psychotherapy side effects. World Psychiatry: Official Journal of the World Psychiatric Association, 13, 306–309. doi:10.1002/wps.20153

21.

Mohr

D. C.

(1995). Negative outcome in psychotherapy: A critical review. Clinical Psychology: Science and Practice, 2, 1–27.

22.

Moritz

Fieker

Hottenrott

Seeralan

Cludius

Kolbeck

. . . Nesoriuc

(2015). No pain, no gain? Adverse effects of psychotherapy in obsessive-compulsive disorder and its relationship to treatment gains. Journal of Obsessive-Compulsive and Related Disorders, 5, 61–66.

23.

Nemirovski Edlund

Carlberg

(2014). Psychodynamic psychotherapy with adolescents and young adults: Outcome in routine practice. Clinical Child Psychology and Psychiatry. Advance online publication. doi:10.1177/1359104514554311

24.

Nutt

D. J.

Sharpe

(2008). Uncritical positive regard? Issues in the efficacy and safety of psychotherapy. Journal of Psychopharmacology, 22, 3–6. doi:10.1177/0269881107086283

25.

Olatunji

B. O.

Deacon

B. J.

Abramowitz

J. S.

(2009). The cruelest cure? Ethical issues in the implementation of exposure-based treatments. Cognitive and Behavioral Practice, 16, 172–180.

26.

Parker

Fletcher

Berk

Paterson

(2013). Development of a measure quantifying adverse psychotherapeutic ingredients: The Experiences of Therapy Questionnaire (ETQ). Psychiatry Research, 206, 293–301.

27.

Rozental

Boettcher

Andersson

Schmidt

Carlbring

(2015). Negative effects of internet interventions: A qualitative content analysis of patients’ experiences with treatments delivered online. Cognitive Behaviour Therapy, 44, 223–236.

28.

Strupp

Hadley

S. W.

(1977). A tripartite model of mental health and therapeutic outcomes: With special reference to negative effects in psychotherapy. American Psychologist, 32, 187–196.

29.

Tan

J. O.

Fegert

J. M.

(2004). Capacity and competence in child and adolescent psychiatry. Health Care Analysis, 12, 285–294.

30.

Warren

J. S.

Nelson

P. L.

Mondragon

S. A.

Baldwin

S. A.

Burlingame

G. M.

(2010). Youth psychotherapy change trajectories and outcomes in usual care: Community mental health versus managed care settings. Journal of Consulting and Clinical Psychology, 78, 144–155.