Send email for updates
Hill C. E., Knox S., Pinto-Coelho K.G.
Psychotherapy: 2018, 55(4), p. 445–460.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Hill at firstname.lastname@example.org.
It’s a dilemma for all therapists and counsellors – how much to disclose about yourself. Another difficult decision is when to directly confront what is happening and being felt then and there in therapy. A review for the American Psychological Association finds that positive reactions generally follow these therapist interventions, but they can also backfire. Guidance is offered on when to try them.
Summary [Though not specific to clients with drug and alcohol problems, the principles derived from this review of psychotherapy studies are likely to be applicable, partly because severe substance use problems generally form part of a complex of broader psychosocial problems. Addictions work may not necessarily best be conceptualised as psychotherapy, but there is a therapeutic element to it which makes these findings relevant to keyworkers and counsellors.]
Commissioned by a task force of the American Psychological Association, this review amalgamated findings relating client reactions during therapy to times when therapists disclosed personal information (‘therapist self-disclosure’) or directly addressed what is happening and being felt at that time (‘immediacy’).
Therapist self-disclosure and immediacy were mainly followed by positive reactions, indicative of improved therapy relationships and mental health functioning, and gaining insight.
The analysis could not establish that self-disclosure and immediacy helped cause the observed client reactions, but the pattern of results was consistent with this possibility, as were the findings of a study which deliberately altered the level of therapist self-disclosure. In turn, it seems likely that these therapist actions affected ultimate outcomes.
The featured review is one of several in a special issue of the journal Psychotherapy devoted to features of the therapist-client relationship related to effectiveness, based on the work of a task force established by the American Psychological Association. This particular review examined the links between the therapist revealing personal information about themself (‘self-disclosure’) and immediately subsequent positive or negative (in therapeutic terms) reactions by psychotherapy clients. A similar analysis was done for therapist comments or questions about what is going on then and there in the therapy session or how they or the client are feeling, termed ‘immediacy’. Unfold the supplementary text for fuller definitions and examples.
Self-disclosures tend to be brief and not generate further discussion, whereas immediacy tends to involve several exchanges as therapist and client discuss and process their feelings about their relationship. In practice, self-disclosure is rare, in one study, accounting for just 1% of all therapist responses; immediacy is much more common, accounting in studies for 5% to 38% of therapist responses.
Therapist self-disclosure and immediacy have long been controversial. Psychoanalysts were traditionally urged to be ‘blank screens’ on to which clients could project feelings and perceptions, though more recently some have suggested therapy can be facilitated by therapists self-disclosing and talking about their relationship with the client. Having long advocated therapist transparency and genuineness, humanistic theorists see therapist self-disclosure and immediacy as curative elements, while cognitive therapists have seen these as helping to address problems in the therapeutic relationship.
To assess the evidence on these issues the reviewers searched for and found 21 studies of psychotherapy sessions published in English which related therapist self-disclosure and/or immediacy to subsequent events, which if not necessarily in the next exchange, occurred soon within the same session, and were seen by the researchers as having been associated with self-disclosure or immediacy. Instances of self-disclosure or immediacy had to have been identified by trained observers from records of therapy sessions, or by clients or therapists after the session(s). Most studies had been conducted in the USA. It should be stressed that analyses of these studies could not establish what caused what, only what occurs following and may be associated with certain therapist behaviours.
Therapist self-disclosure and immediacy have most often been assessed by trained judges who code these interventions as present or absent in sentences or speaking turns in taped or transcribed sessions, but also by defining these terms to clients or therapists and asking them to report instances in preceding sessions.
A third method involves trained judges listening to an entire session and estimating how frequently or how well the therapist used these interventions. Three widely used measures to facilitate this are: the Multitheoretical List of Therapeutic Interventions; the Psychotherapy Q-Set; and the Comparative Psychotherapy Process Scale.
Given the methodology of most of the relevant studies, conventional quantitative methods could not be used to amalgamate their results. Instead, a ‘qualitative’ meta-analysis was conducted. In practice this entailed first developing ways to categorise how the studies described what followed self-disclosure or immediacy. For example, the category “enhanced therapy relationship” included instances when the tasks of therapy were clarified, boundaries negotiated, when clients had gained new ‘corrective’ understandings as a result of their interaction with the therapist, expressed positive feelings about the therapist, or when ruptures in the client–therapist relationship were repaired. See the panel below for the categories the researchers used to analyse the studies.
Findings were amalgamated in a way which meant each study contributed to the average in proportion to the number of clients studied. For example, if a therapy process was associated with preceding self-disclosure or immediacy once in a single-client case study, it was coded 1/1; if it occurred nine times among 13 clients in a study, it was coded 9/13; if it occurred zero times in a study of 15 clients, it was coded 0/15; and in study which instead reported effect sizes, a medium effect among 30 clients was coded 22/30. Across all four studies, the process occurred among 32 of 59 participants, or 54%. Only if the frequency of a category of therapy process differed by at least 30% from another (eg, among 70% v. 40% of clients) was the difference considered meaningful.
1 Client mental health functioning improved, such as less severe symptoms, improved relationships outside therapy, more positive feelings about oneself, or desired changes in behaviour.
2 Client opened up/explored/experienced feelings.
3 Client gained insight.
4 Client felt understood, that they were not so abnormal, and reassured.
5 Client used immediacy.
6 Overall helpful (nonspecific) for client.
7 Enhanced therapy relationship, including: clarified tasks of therapy; negotiated boundaries; client gained new ‘corrective’ understandings as a result of their interaction with the therapist; client expressed positive feelings about the therapist; or client helped repair a rupture in relationship.
8 Impaired therapy relationship. For example, client felt a lack of clarity about the relationship, role confusion, blurred boundaries, or rupture.
9 Client had negative feelings/reactions.
10 Client openness/exploration/insight was inhibited.
11 Overall not helpful (nonspecific) for client.
12 Negative effects for therapist.
13 Overall neutral reactions/no changes for client.
Therapist self-disclosure and immediacy were most often followed by client comments indicative of therapeutically positive feelings, thoughts, or understandings, termed by the reviewers as therapy ‘processes’. Across the 21 studies of in total 184 clients, the most frequent processes to follow self-disclosure or immediacy were an enhanced therapy relationship (60% of times), improved client mental health functioning (42% of times), the client gaining insight (38% of times), and other unspecified occurrences which seemed helpful for the client (36% of times). The most frequent negative process was unspecified events which seemed unhelpful for the client (26% of times), followed by an impaired therapy relationship (19% of times). Least frequent were inhibited client openness/exploration (6% of times) and negative effects for the therapist (5% of times).
Next, results for self-disclosure and immediacy were analysed separately and compared. Just five studies of in total 99 clients focused on therapist self-disclosure not combined with immediacy. Again, subsequent processes were most often positive. The most frequent to follow self-disclosure were an enhanced therapy relationship (64% of times), the client gaining insight (46% of times), improved client mental health functioning (45% of times), and other unspecified responses which seemed helpful for the client (37% of times). Most frequent adverse processes were negative feelings/reactions by the client (30% of times) and unspecified events which seemed unhelpful for the client (29% of times). Given so few studies with such varied assessments, these results should be viewed with caution.
Fifteen studies of in total 78 clients focused on immediacy as a separate skill. What followed immediacy was mostly positive, but some negative processes were also common. The three most frequent processes were enhanced therapy relationships (40% of times), the client opening up (40% of times), and unspecified events unhelpful for the client (39% of times). However, five of these studies were of positive events only, namely resolution of ‘ruptured’ therapeutic relationships. Just six analysed all events occurring within sessions, both positive and negative. Among these six, what followed immediacy was again mostly positive, including the client opening up (60% of times) and gaining insight (41% of times), but unspecified events unhelpful for the client were also very common (50% of times). Also common (55% of times) was the client themself responding with immediacy.
Comparing what follows therapist self-disclosure versus immediacy was done on the basis of the six immediacy studies which like all five self-disclosure studies, included both positive and negative events. However, other differences between the studies made it difficult to attribute differences to different types of therapist intervention. With this limitation in mind, five meaningful differences were found. Compared to immediacy, therapist self-disclosure was more often followed by improved mental health functioning, unspecified helpful processes, and enhanced therapy relationships, but less often by the client opening up and the client’s use of immediacy. These differences make some sense, because therapists typically self-disclose to facilitate client self-exploration (eg, ‘When I have been in your situation, I felt angry. I wonder if you feel that way?’) with the aim of fostering understanding between them and improving mental health functioning. In contrast, immediacy often focuses on the client–therapist relationship, so the resulting processes are more often collaborative and focus on both participants (eg, ‘You mentioned not feeling respected in relationships, and I’m wondering how you’re feeling about our relationship?’).
Subsequent processes associated with therapist self-disclosure and immediacy were largely positive, though the differences suggested that self-disclosures were the more helpful, supportive interventions. In contrast, immediacy was more likely to be associated with clients opening up and responding with immediacy, suggesting they can be useful micro-interventions for dealing with problems in the therapeutic relationship. However, clients can react negatively when therapists talk about their personal lives or relationship with the client. Therapists too can feel vulnerable and incompetent when they attempt these manoeuvres. In the reviewed studies, negative processes were seen in up to 30% of cases.
Small numbers of studies and of participants mean these findings are tentative. Also, despite the close relationship in time between therapist self-disclosure/immediacy and subsequent processes, causality cannot be assumed because just one of the studies manipulated self-disclosure or immediacy to test whether these really were active ingredients in causing the observed processes.
Nor can it be said on the basis of these studies that self-disclosure and immediacy are distinct in being followed by the observed processes; might also follow other micro-interventions such as interpretations and reflections of feelings. It is important to recognise that therapists do not use self-disclosure or immediacy on a random basis, but rather for a specific purpose in a specific context; it cannot be said that self-disclosure or immediacy are positive strategies no matter what the situation, just that they may be when selected by therapists for certain purposes in certain contexts. There were also other obstacles (unfold supplementary text) to interpreting the findings.
Because they are most often followed by positive processes, therapists might consider using self-disclosure and immediacy. In practice these are used infrequently, in line with reviews and guidelines which stress using them sparingly and deliberately.
In general the findings suggest that effective self-disclosures focus on clients and use the therapist’s experiences to facilitate client exploration, which fosters improved understanding and functioning. Therapists might especially consider disclosing when clients feel alone, vulnerable, and in need of support; for clients to learn that are not the only ones who have felt this way can generate a sense of the universality of such experiences.
In contrast, therapists often use immediacy to negotiate and address problems in the relationship. Consequently, therapists might consider using immediacy primarily to help clients open up and talk about underlying feelings, especially when negotiating the therapeutic relationship. However, talking about the relationship can lead to volatility as problems are illuminated, so therapists will need to be aware of, open to, and prepared to address their own reactions and those of the client.
Integrating the findings of the featured review with those of the broader literature suggests the following specific practice recommendations, firstly with regard to therapist self-disclosure:
• Be cautious, thoughtful, and strategic about using self-disclosure and use sparingly.
• Have a client-focused intention for using self-disclosure and make it relevant to comments or issues raised by the client.
• Evaluate how clients might respond and whether self-disclosure is likely to help them.
• Make sure the therapeutic relationship is strong before using self-disclosure.
• Keep the disclosure brief with few details.
• Disclose issues you have resolved rather than those which remain unresolved.
• Focus on similarities between therapist and client.
• Focus on the client’s rather than the therapist’s needs.
• After self-disclosure, turn the focus back on the client.
• Observe the client’s reactions to the self-disclosure, assess effectiveness, and decide whether it will be appropriate to use therapist self-disclosure again.
Recommendations on immediacy are:
• Be aware that immediacy often involves lengthy processing.
• If therapists want clients to be immediate, they should be immediate with their own feelings.
• Be attentive to how the client responds to immediacy; many clients are not comfortable with it, and it is sometimes associated with negative effects.
• Be aware of and examine the possibility of countertransference (when a therapist’s internal conflicts affect reactions to patients) and consult with others to ensure you are acting in the client’s best interests.
commentary The reviewers stressed that their analysis could not establish whether therapist self-disclosure or immediacy led to the reactions they catalogued in the clients, and made no attempt to link therapist behaviour to the outcomes of therapy. Nevertheless their practice recommendations were based on assumptions that modest levels of these behaviours can help the client – assumptions which seem justified, if in need of confirmation.
A fundamental limitation was that the analysis did not quantify how often the same types of client comments occurred in the absence of self-disclosure or immediacy. It could be that these comments emerged out of the flow of the conversation, and would have happened anyway. Arguing against this, is that it ‘makes sense’ for some of the types of comments to be prompted by self-disclosure or immediacy, and that in comparable studies, the types which followed each were often very different. For example, just 9% of clients exhibited improved mental health functioning following immediacy, but 45% after self-disclosure, while corresponding figures for the client opening up were 60% and 27%, and for client immediacy, 55% and 0%. With such large differences which also ‘make sense’, it seems likely that these therapist behaviours often did help prompt client responses which would not otherwise have occurred. If this is the case, unless we believe what happens during therapy makes no difference to its outcomes, it becomes likely that these micro-interventions also influence what in the end the client gets out of therapy. These extrapolations are supported by a key study described below.
Even an analysis of the kind suggested above (comparing what happens after self-disclosure or immediacy with what happens after other therapist behaviours) would not have been enough to establish causality, because so many other influences would have been in play. To most securely establish causality requires deliberately manipulating the degree of self-disclosure or immediacy while keeping all else the same. Only one of the reviewed studies attempted this, specifically with self-disclosure. It confirmed that self-disclosure does affect the therapeutic interaction and the client, and suggested that modest levels of reciprocal self-disclosure by the therapist in response to the client are not harmful and likely to be beneficial. Details below.
Published in 2001, the study was conducted at a US university therapy/counselling clinic also open to the general community. For the study, treatment was offered in four weekly sessions but could continue outside the remit of the study. Recruited to the study were 36 adult clients whose presenting problems included depression, social or performance anxiety, relationship conflicts, or lack of impulse control. In pairs they were randomly allocated to 18 therapists and within each pair also at random to either increased or decreased therapist self-disclosure. Therapists knew the purpose of the study but were given plausible rationales for both using and restricting disclosures. Then at random they were told with one of their clients to increase the frequency with which they disclosed personal information in response to similar client self-disclosures, and to match the level of language, sophistication and intimacy of the client. For their other client they were told to restrict disclosure of personal information and to maintain the focus on the client, responding to client self-disclosures in other ways (examples were given).
Following each session clients rated how distressed they were by each of a list of symptoms, how much they liked the therapist, and how often the therapist had self-disclosed to them. Audiotapes of the sessions were also rated by trained observers for the frequency and intimacy of therapist and client self-disclosures as well as the duration of therapist self-disclosures. Also rated were whether these were judged to be a response to similar client concerns; not surprisingly given the instructions to the therapists, virtually all were. For each client all these assessments were averaged across the four treatment sessions.
Clients and observers agreed that, as intended, therapist self-disclosures were more common when therapists had been instructed to increase the number, and observers also rated them as longer and more intimate. Having successfully altered self-disclosure levels, how clients reacted was the crucial issue; the results suggested self-disclosure had positive effects. Compared to low-disclosure instructions, when therapists had been instructed to more often self-disclose, clients said they felt less symptom distress and liked their therapist more. These findings are in line with those of the featured study that among the most frequent therapy processes to follow self-disclosure were an enhanced therapy relationship (64% of times) and improved client mental health functioning (45% of times).
Importantly, these results were obtained even though when instructed to self-disclose more, these events remained rare – on average four to five per session each lasting less than 15 seconds. This level of therapist self-disclosure did not affect how often or how intimately clients self-disclosed.
The findings were considered to confirm theoretical expectations that self-disclosure by the therapist can enhance the therapist–client relationship and be helpful for treatment. However, this was demonstrated only in relation to ‘reciprocal’ therapist self-disclosures which were a response to, or similar to, issues first raised by the client. More gratuitous or more extensive self-disclosures may not have had the same effects, and could be negative if they detracted from the focus on the client.
Listed below are analyses of the other reviews commissioned by the American Psychological Association task force.
Overall conclusions based on the other 16 reviews
The ‘real’, person-to-person relationship
Alliance in couple and family therapy
Alliance in child and adolescent therapy
Cohesion in group therapy
Goal consensus and collaboration
Therapist and client emotional expression
Repairing ruptured alliances between therapists and clients
Treatment outcome expectations
Feeding back client progress data therapists
Last revised 28 March 2019. First uploaded 12 January 2019
Give us your feedback on the site (two-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates
REVIEW 2018 Congruence/genuineness: a meta-analysis
REVIEW 2018 Psychotherapy relationships that work III
REVIEW 2018 Cohesion in group therapy: a meta-analysis