Can Therapists and Patients Be Friends? Ethics of Post-Therapy Contact
by Sam Brandsen
“Can therapists be friends with their patients?”
The most common answer I’ve seen to this question is that friendships between therapists and former patients raise significant ethical issues. After all, the power dynamic between a therapist and patient certainly doesn’t dissipate the moment that a friendship is declared, and the patient may be left in a very isolating and vulnerable situation within an attempted friendship. One particularly interesting study [Kramer, 2016] involved interviewing 9 individuals who had a friendship with their previous therapist and found that confusion about the therapy-friendship relationship was a central theme in all interviews. Other ethical guidelines (e.g., APA and ACA codes of ethics) also emphasise that dual relationships can create lasting harm.
This has prompted me to think about the following questions:
Are there any hypothetical “edge cases” where friendship could ever be feasible between a therapist and former patient/client?
How (if at all) should considerations about a therapist and patient maintaining regular contact post-therapy change in situations that are already complicated in some way? Some example situations include:
Therapy needing to end unexpectedly or suddenly due to external stressors or events in the therapist’s life or other external factors
Cases where boundaries have already become significantly blurred in therapy, such that recovering a relatively healthy therapeutic relationship may be unlikely or impossible
A therapist and former client finding themselves unexpectedly in the same environment (course, workplace, other organisation) post-therapy
What supports are there for clinicians who are navigating therapy termination and potentially making complicated decisions about where to draw the line regarding post-therapy contact with former patients?
This below is my attempt to touch on some of these questions. I personally believe that the risk of harm in attempting a friendship with a patient is far too high and that a true friendship – or any kind of equal/mutual relationship – is likely impossible given the power dynamics of therapy. However, I also suspect that there are situations where maintaining some form of appropriate ongoing communication after therapy (e.g. a former patient periodically emailing their therapist updates) may be less harmful than an absolute termination of contact; especially if this communication is approached in a thoughtful way with adequate consultation and supervision. To this aim, the end of this document includes a tentative set of guiding questions for clinicians
Major Disclaimer I am not a clinician and, while I am a researcher, I haven’t completed any formal research on this topic. I have some relevant personal experience as a previous patient; but of course my own experience is only one data point and there is so much that I still have yet to learn about clinical ethics, risk of clinical harm, etc.
I view the below as a first iteration of potential ideas on this topic and hope to refine them over time based on any feedback or experiences that others would like to share.
Additionally, the thoughts below are my own, but due to how incredibly long my initial writing was (and my own challenges in making things more concise!) I used ChatGPT to condense and summarise my initial essay.
Case Studies
Case #1: Single, brief session years ago
Scenario:
A therapist and patient met for a single session over decade ago. The session was not particularly memorable—nothing went especially well or badly. The patient did not disclose anything that was particularly sensitive (or that would be sensitive to them now), and ultimately ended up seeing another therapist who was a good fit.
The patient and therapist have now met again in a completely different setting (e.g. shared hobby group).
Impression:
This is the only scenario I can imagine where there is at least a solid possibility that a friendship may be possible without excessive risks
It is still probably beneficial for the therapist to avoid initiating a friendship and to have a direct discussion with the former patient before accepting any kind of regular social interactions. In some cases, the previous therapy session may have had a more pivotal impact on the patient than the therapist realized and there is a chance that the patient still (understandably) views them in a therapy-like role. Additionally, if the patient has had traumatic or difficult experiences with other providers, they might be especially sensitive to further harm from even a very distant clinical figure.
However, excessive focus on a potentially distant power dynamic may not always be helpful, especially if the memories are so distant that the patient could barely remember even seeing the therapist.
In some cultural or community contexts (e.g., rural towns), overlapping roles may be more common or inevitable. In these situations, the question may become how boundaries can be renegotiated with cultural sensitivity/sensitivity to the community context while protecting patient wellbeing.
Case #2: Warm, ongoing therapy relationship
Scenario:
A therapist and patient have a therapy relationship with warm rapport and relatively intact boundaries. The patient enjoys their time in therapy and expresses that they wish the therapist could be their friend.
Impression:
Being friends in this scenario is very risky for the patient.
The patient may be wanting a friendship that extends parts of the therapeutic relationship (unconditional care, an attentive listener, someone deeply attuned to their thoughts and prior experiences) into their “daily life”. While this is an extremely understandable desire, the patient may struggle with the realities of an attempted friendship.
This is not to imply that the patient cannot genuinely have a human appreciation for certain traits or qualities or even unique shared interests that the therapist possesses, but in a typical therapy relationship, the patient does not have enough information about what a friendship would truly look like to make an informed decision.
They know how their therapist interacts with them as a therapist, but probably do not know the therapist’s availability, personal needs, or communication style as a friend.
Key Points: Consent and Countertransference
While it is important to hear patients’ wishes, there are things patients simply cannot consent to. Patients may sincerely want friendship, ongoing closeness, or even deeper relationships with their former therapist; but this does not erase the power dynamic, the risks of blurred roles, or the therapist’s duty of care. A documented record that “the patient asked for this” does not protect against harm, nor does it absolve the therapist of responsibility. True consent in this context is limited.
With that said, therapists can certainly still acknowledge their patient’s feedback even in situations where it wouldn’t be okay to apply the feedback literally. For example, suppose the patient indicates that part of the reason they wish to be friends is because they feel insecure about whether they are likable enough/a good enough “candidate” for friendship— and having their therapist affirm that through friendship would feel validating. In this case, the therapist might explicitly highlight moments in therapy where the patient has shown qualities (warmth, humor, thoughtfulness, persistence) that would make them a good friend to others, while still maintaining the boundaries of the clinical relationship.
Likewise, it is critical for therapists to be aware of their own countertransference in order to avoid (likely inadvertently) selectively using feedback or communication from their patients in a way that feeds their own wishes. Namely, the therapist’s feelings of grief, guilt, affection, or professional insecurity during termination may become a primary driver of blurred boundaries. This means that protecting patients often requires therapists to address their own needs first, ideally through consultation or supervision.
Case #2b: Therapy ending due to therapist life changes
Scenario:
Same as Case #2, but the therapist is retiring or undergoing major life changes/illness and cannot continue therapy in the way the patient needs. As such, the therapist wonders if offering a friendship may help offset the stress of therapy termination.
Impression:
This is a tricky situation. Countertransference may also apply here – e.g., a therapist retiring might themselves feel loss, guilt, or longing for continued connection, which could make consideration of friendship inadvertently more about their own needs than the patient’s.
A friendship might paradoxically carry more risks, as the therapist may be under additional stress and the patient could be brought into a friendship dynamic during a turbulent time. This may lead to the patient quickly becoming a source of support for the therapist, in an unhealthy role reversal. (Additional risks are discussed after Case 3)
On the other hand, sudden termination or transferring the patient to another therapist may also be stressful, and a strict dichotomy of “termination and no contact” versus “switching to a friendship” could push therapists to prolong therapy unnecessarily, even if it’s not in the patient’s best interest.
Potential Strategies:
Instead of attempting a friendship, clinicians could consider carefully structured methods of ongoing communication. Friendship implies an equal, mutual social relationship, while ongoing connection refers to carefully structured, limited forms of post-therapy contact (e.g., symbolic tokens, update emails, mediated check-ins).
Ideally, this process would be:
Opt-in: instead of putting patients on the spot with questions about ongoing communication, consider having an exit feedback form that is distributed to all patients, where patients can select if they would like to consult about ongoing communication.
Determined with consultation (e.g., with another clinician) to guide decision-making, while ensuring the patient also has access to consultation. This reduces the chance of making exceptions in isolation and provides oversight and accountability.
Focused on preserving elements of care for the patient in a controlled way, even if the traditional therapy relationship cannot continue. For example, if a patient’s main wish is to have some kind of reminder of their time in therapy, a letter or symbolic token might be beneficial. If the patient feels stress about no longer being able to share their milestones with their clinician, perhaps structuring a way for occasional updates could ease part of that stress.
Case #3: Blurred boundaries and intense attachment
Scenario:
Now suppose that therapy has already become somewhat – or quite a bit – like a friendship, possibly due to dual roles, social overlap, extensive self-disclosure, or other factors.
The patient may have developed an intense attachment and may feel switching to a friendship is a more accurate reflection of the relationship. They may also feel stuck, potentially being afraid that ending therapy would mean losing the connection with their therapist entirely.
Given the level of blurriness already existing, the therapist may have been using the relationship to meet some of their own emotional needs, even without realizing it, or otherwise may have personal factors influencing their decision making. The therapist may wonder if switching to a friendship could be a form of “harm reduction” by providing a setting where the blurred boundaries would be less clinically concerning.
Impression:
This is a difficult situation and the patient may already be impacted by blurred boundaries, even if they haven’t yet consciously identified that they are impacted.
Friendship may feel less risky because both parties feel familiar with the relationship, but the risk of harm is very high and the complex situation may heighten confusion for the patient.
Patient Risks:
Confusion about the relationship: Uncertainty over whether interactions are therapeutic or social.
Increased attachment and dependency: The patient may rely emotionally on the therapist-turned-friend, expecting the same level of support previously provided in therapy. Likewise, the patient may find the relationship to be quite intense (blurring unconditional support and understanding of therapy with the closeness of friendship) in a way that ultimately isn’t healthy for them or that limits them from more fully engaging in other relationships of their life.
Anxiety and uncertainty: Questions like “Is it still okay to talk about my mental health? How much? Will she feel I’m still treating her as a therapist?” can arise, making the friendship feel overwhelming or ambiguous.
Difficulty adjusting to changes: Transitioning from therapy to friendship may trigger stress over how often to meet, whether virtual vs. in-person is acceptable, and how to offer support appropriately.
Emotional impact of disagreements or boundary enforcement: Even minor conflicts may feel amplified for the patient, because the patient may perceive the therapist as still having a partially clinical, “objective” role. Additionally, the therapist may – even if unintentionally or indirectly– use their knowledge of the patient’s life, trauma, and vulnerabilities in a moment of anger or frustration.
Guilt over having expectations, fear. Alternatively, the patient may be hyper-aware that they are now in a less “protected” or unconditional relationship, and may worry that their therapist would drop them as a friend at any moment if they are not a “good enough” or entertaining enough friend.
Struggle with autonomy and decision-making: The patient may feel responsible for the therapist’s feelings or well-being, mirroring dynamics from the therapy relationship.
Exacerbation of prior trauma or vulnerability: The ambiguous role of the therapist/friend may trigger unresolved attachment issues or trauma-related responses.
Therapist Risks:
Attachment and anxiety about loss: By the time therapy has blurred into friendship-like dynamics, the therapist may themselves be reluctant to end the relationship. They may fear losing the patient’s presence, feedback, or affection.
Identity and professional insecurity: Some therapists may unconsciously rely on their patient’s admiration or closeness to feel competent or valued. The threat of termination can provoke anxiety about being “replaceable” or “not good enough.”
Fear of rejection in friendship: If the patient doesn’t like the “non-therapist” version of them, the therapist may feel hurt or betrayed — and unlike in therapy, they may not feel they have a structured way to process or contain those feelings.
Defensiveness and external pressure: Therapists may fear for their license, or for how colleagues or supervisors would judge them if the blurred boundaries came to light. This can lead to defensive minimization (“this was all a clinical decision”) rather than open acknowledgment of their own needs, which can be invalidating or destabilizing for the patient.
Unacknowledged reciprocity: Even if therapists consciously frame interactions as “for the patient’s benefit,” they may be unconsciously seeking emotional support, affirmation, or even professional favors — which complicates the power dynamic further.
Resentment. If the friendship ends up being less ideal than hoped (e.g. the therapist doesn’t actually enjoy their client as a friend or the patient needs a lot of support processing the switch to a friendship), the therapist may begin to feel resentful and may begin to view interacting with the patient as an obligation
Risk of isolation: Because admitting to blurred boundaries can feel taboo, therapists may avoid consultation and instead make decisions alone, amplifying the likelihood of defensive, guilt-driven, or self-protective choices.
Potential Mitigation:
Seek consultation with another clinician and/or mediator support for the patient.
Transparent discussions about expectations, boundaries, and the potential impact of friendship, while keeping in mind limitations on what patients can realistically consent to.
Highly structured, limited interactions focusing on patient needs.
Ongoing therapist self-reflection to prevent decisions driven by guilt, attachment, or avoidance of termination discomfort.
General Principles
Attachment intensifies risk: Warm, ongoing, or blurred therapeutic relationships make friendship potentially harmful (Cases #2–3).
Consultation and supervision are essential for ethical practice.
Therapist self-awareness: Ambivalence, guilt, anxiety, or isolation can influence decisions and exacerbate risks.
Patient vulnerability: Prior trauma, dependence, or attachment must be central considerations.
Therapist countertransference is a recurring factor: Even in cases where risks seem low, therapists may carry their own needs (desire for connection, affirmation, or relief from guilt) that influence decision-making. Recognizing and addressing this is essential across scenarios, not just when boundaries have already blurred.
Guiding Questions for Therapists
Patient Factors
Has the patient had significant attachment to me as a therapist?
Could this post-therapy connection confuse or exacerbate their trauma, attachment patterns, or sense of autonomy?
Would this connection risk limiting their capacity to build other relationships?
Am I truly considering the patient’s vulnerability, or am I taking their stated preferences at face value without weighing the underlying power imbalance, attachment needs, or mental health factors?
Therapist Factors
What am I personally wanting from this connection (companionship, affirmation, relief from guilt, avoidance of grief over termination)?
Am I feeling pressure to say yes out of fear of hurting the patient, guilt, or fear of judgment?
Would I feel comfortable discussing this decision openly with a supervisor or colleague?
Structural / Practical Factors
Do I have the bandwidth to sustain the type of connection the patient may reasonably expect?
Would this connection create obligations I cannot realistically fulfill (e.g., frequent emotional support, availability during crises)?
If the connection ends badly, could it leave the patient worse off than a clean termination would have?
Fairness and Consistency
Am I making an exception for this patient, or would I consider the same with any patient?
Does this decision respect fairness across my practice, or am I privileging one relationship in a way that could be harmful to others (or to my own boundaries)?
Caveat: If I have already made significant exceptions with this patient, “reverting” to typical professional boundaries may feel abrupt, confusing, or invalidating. In these cases, fairness may not be fully achievable, and my responsibility is to manage the transition with clarity, acknowledgment of the past exceptions, and a gradual process where possible.
Risk Mitigation
Can this connection be structured in a way that limits harm (time-limited, topic-focused, occasional check-ins, hobby-based)? Alternative ways of connection could include:
Indirect connection (e.g. the therapist becomes connected the patient’s new therapist or another consultant who can help transfer periodic well-wishes or other communication) from the therapist to the patient
A small token of connection (could be anything – a rock, a journal, something else with a symbolic gesture) that the patient can use to feel some kind of connection with their previous therapist even if the relationship has to terminate
A limited, joint project (e.g. therapist and patient both periodically contribute to a private classical music playlist on Spotify or something– would need to take care not to introduce new dual roles with this)
Structured way for patient to send life updates to their therapist with clear expectations about how the therapist can respond
Ongoing, occasional meetings facilitated by a third party
Other
Do both the patient and I have external support systems to process this transition?
Have I sought consultation to ensure I am not making this decision in isolation?
Deeper Reflection / Pause Signals
How did I arrive at the point of even considering friendship with this patient — was this shaped by blurred boundaries, countertransference, or unprocessed dynamics within therapy?
Has there been potential harm from therapy itself (e.g., excessive self-disclosure, dependency, unmet needs) that needs to be identified and addressed before considering any next steps?
Am I prepared to respond non-defensively if the patient later feels harmed by this connection, and to take responsibility for my role?
Could I tolerate the possibility that the patient may reject the social relationship at any point once it begins or feel ambivalent about it?
Have I fully considered whether there are less risky alternatives — such as a gradual, supportive termination, periodic structured check-ins, or a mediated form of ongoing contact?
Reference:
[Kramer, 2016] Allison Kramer. Why can’t we be friends? American Counseling Association, 2016. https://www.counseling.org/publications/counseling-today-magazine/article-archive/article/legacy/why-cant-we-be-friends
