A practical guide to understanding extended client talk and helping sessions remain focused and useful
Most counsellors have experienced a session in which the client begins talking and barely pauses for the next 40 minutes. Several important moments appear along the way, but before the counsellor can respond, the client has moved to another event, another person or another part of the story. By the time there is an opening, the session is nearly over.
The counsellor may leave feeling ineffective, crowded out or unsure whether any therapeutic work occurred. The client may leave relieved to have spoken, disappointed that nothing was resolved, or perfectly satisfied with the session. Sometimes the counsellor does not know which.
Calling someone a “talkative client” does not tell us very much. Extended talking can reflect loneliness, anxiety, a need to feel understood, a verbal approach to organising thoughts, repetitive rumination, difficulty with conversational regulation, or an attempt to avoid emotion. It can also be ordinary enthusiasm, a cultural communication style or the first opportunity someone has had to tell an important story in full.
The amount a client talks is therefore a poor basis for deciding what to do. The more useful question is:
What is the talking doing for this client, and what happens when the counsellor tries to slow, focus or interrupt it?
First decide whether the talking is actually a problem
There is no evidence-based universal ratio of client-to-counsellor speaking time that defines a productive session. Clients generally need enough opportunity to describe their experiences in their own terms, and therapist empathy is reliably associated with better psychotherapy outcomes (Elliott et al., 2018). A client speaking for much of a session is not necessarily evidence that the session has gone wrong.
At times, being listened to is the work. This may be particularly true when someone is disclosing an experience for the first time, has recently experienced a crisis, or has rarely been able to tell their story without being doubted, corrected or hurried.
Extended talking becomes more concerning when it repeatedly prevents the client and counsellor from working on the client’s stated goals. You might notice that:
the client brings several important subjects into each session but cannot remain with any of them;
the same problem is described repeatedly without any development in understanding, emotion, decision-making or action;
the client becomes distressed when the counsellor speaks, redirects or respectfully interrupts;
the counsellor regularly disengages, loses track or stops attempting to contribute;
important clinical information is missed because the counsellor cannot ask questions or clarify;
endings are consistently rushed, with significant new material introduced in the final minutes;
the client says that sessions are not helping, despite using most of each session to report what has happened since the last appointment.
Before introducing more structure, ask the client what they find useful about the current pattern. The counsellor’s sense that “nothing happened” may not match the client’s experience. It is also worth asking whether the client wants primarily to be heard, to understand a pattern, to make a decision, to learn a skill or to work towards a particular change. Different purposes require different conversations.
Why might a client talk at length?
They have few other opportunities to talk
For some people, counselling is one of the only regular conversations in which another person gives them sustained attention. This may be especially relevant for clients who live alone, have lost important relationships, are caring for others, or have relationships in which there is little interest in their internal experience.
Emotional experiences commonly create an urge to tell other people what happened. Rimé (2009) described this as the social sharing of emotion. People share emotional experiences to seek comfort, recognition, understanding, social connection and help making sense of what occurred.
Loneliness also involves a desire to reconnect, although it can heighten sensitivity to rejection and make social interactions more difficult (Hawkley & Cacioppo, 2010). There is no evidence that loneliness inevitably makes clients talk more. It is nevertheless a reasonable hypothesis when a client arrives with a week’s worth of conversation stored up and little opportunity to share it elsewhere.
In this situation, simply restricting the talking may repeat the experience of having nobody interested enough to listen. The counsellor can acknowledge the client’s need for conversation while also examining whether the client wants help developing more contact and reciprocity outside therapy.
They have not yet felt understood
Clients sometimes repeat themselves because the counsellor has responded to the facts but missed the central emotional meaning. A person may provide more examples, further evidence and increasingly detailed explanations in an attempt to produce an experience of being accurately understood.
Research on interpersonal responsiveness suggests that disclosure contributes to intimacy partly when the listener is experienced as understanding, validating and responsive (Laurenceau et al., 1998). Validation can also affect emotional intensity, although its effects vary according to the emotion and the person’s level of emotion dysregulation (Kuo et al., 2022).
When this is the function of the talking, another broad invitation such as “Tell me more” may prolong the account without meeting the need underneath it. A concise reflection may be more useful:
“You have explained several versions of what happened. I wonder if the part you most need me to understand is that you were left dealing with the consequences, while everybody else behaved as though it was a minor misunderstanding.”
You can then ask:
“Have I understood the important part, or is there something I am still missing?”
If the client settles, pauses or becomes more emotionally present, the talking may have been partly an effort to reach that point of recognition.
Talking helps them organise their thinking
Some clients work out what they think by speaking. They may begin a sentence without knowing where it will end, make connections as they talk and only identify the important point after providing considerable context.
The familiar term “external processor” describes this experience reasonably well, but it is not a formally established personality category. Research on private and inner speech does support the broader idea that language contributes to attention, working memory, problem-solving and the regulation of thought and behaviour (Alderson-Day & Fernyhough, 2015).
A client who thinks aloud may benefit from structure that helps them retain the useful parts of their own thinking. The counsellor might write down the emerging themes, offer regular summaries, or ask the client to identify the sentence they most want to remember.
For example:
“There have been several important thoughts in what you’ve just said. I’m going to summarise them so we don’t lose the connections you’ve made.”
This does not require the client to arrive with a perfectly organised account. It gives the conversation enough scaffolding for both people to work with it.
Talking regulates anxiety
Some clients increase their pace and volume of speech when they feel anxious. Silence may intensify self-consciousness, uncertainty or awareness of the counsellor’s attention. Continuous talking can prevent the discomfort of waiting to see what the other person will say.
Worry is often predominantly verbal and can function as a way of avoiding more emotionally evocative imagery or sudden changes in emotional state (Newman & Llera, 2011). A client can therefore talk extensively about a distressing subject while remaining relatively distant from the experience itself.
Signs that this may be occurring include a noticeable increase in speed when emotion appears, rapid movement into explanation after an experiential question, or repeated use of detail at moments when the conversation begins to slow.
The counsellor can name the change without making an immediate interpretation:
“I noticed that when we came close to what you felt in that moment, your talking became much faster. What happened for you just then?”
Or we might offer a brief pause:
“I’d like to slow us down for a moment, so we can notice what’s happening for you as we talk about this. Would that be okay?”
A pause should have a clear purpose. Dropping into silence without explanation may feel unnecessarily exposing to someone who already experiences silence as threatening.
The client is ruminating aloud
Repetitive talking can be an interpersonal form of rumination. The client revisits the same problem, speculates about motives, rehearses unfairness and considers multiple hypothetical outcomes. This may feel active because many words are being used, even though the thinking remains in the same place.
Rumination is associated with more persistent negative affect, impaired problem-solving and a greater likelihood of depression and anxiety (Watkins & Roberts, 2020). Rose (2002) used the term co-rumination to describe extensive, repeated discussion of problems within relationships. Co-rumination can increase feelings of closeness while also being associated with emotional difficulties.
Therapy is not an ordinary friendship, and the concept cannot be transferred wholesale. It does remind us that a conversation can feel supportive while strengthening repetitive problem-focused thinking.
When you suspect rumination, listen for whether the account is producing new information or simply another circuit around the same material. It can help to shift from abstract questions about why something happened to concrete questions about what occurred, what the client noticed and what they did next.
You might say:
“We have considered several explanations for why she behaved that way, but none of them can be confirmed. What happens to you when you spend this much time trying to solve her motives?”
Or:
“What do you understand now that you did not understand 15 minutes ago?”
These questions make the process visible without ridiculing the client’s need to think.
Talking keeps the client away from emotion
Detailed storytelling, analysis and explanation can all protect a client from emotional contact. The counsellor may learn everything about the sequence of events while learning very little about what the experience meant to the client.
This does not mean that detail is inherently avoidant. Some experiences require careful chronology, particularly when the client has been confused, gaslit, traumatised or repeatedly told that their account is inaccurate. Avoidance is more likely when the level of detail consistently increases at the point where emotion, vulnerability, uncertainty or relational contact begins.
The counsellor can shift the level of the conversation:
“I understand what happened next. I would like to go back to the moment when he said that. What did you notice in yourself before you began explaining it?”
Another useful intervention is:
“If we set aside the reasons for a moment, what was painful about it?”
A tentative reflection of feeling may help bring the emotional meaning into focus, while leaving the client room to correct the counsellor’s impression.
“As you describe it, I’m wondering if you felt quite alone in that moment. Does that fit, or was it something else?”
If the client cannot answer, becomes more activated or immediately returns to facts, that response is clinically useful. The next task may involve building enough emotional awareness or regulation for closer contact with the experience.
The talking reflects neurodivergent communication
ADHD can affect conversational inhibition, topic maintenance, narrative organisation and turn-taking. A systematic review and meta-analysis found substantial group-level differences in pragmatic language among people with ADHD, although the studies do not support making assumptions about an individual client from talkativeness alone (Kessler & Ikuta, 2023).
Autistic clients may also communicate in ways that differ from non-autistic conversational expectations. The double-empathy framework proposes that communication difficulties between autistic and non-autistic people are partly produced by a mismatch between communication styles rather than by a deficit located within the autistic person (Milton, 2012). Research showing effective information transfer between autistic people provides some support for this relational account (Crompton et al., 2020).
Subtle hints, changes in posture and attempts to “look like” you want a turn may be ineffective and unfair. More explicit agreements can be helpful:
“I sometimes find it difficult to tell when you have finished a thought. Would it be all right if I raise my hand slightly when I want us to pause?”
Other practical adaptations include:
keeping a visible list of subjects raised during the session;
agreeing on a direct signal for redirection;
telling the client how much time remains;
asking one question at a time;
allowing the client to send or bring a short written list;
checking whether the client prefers interruptions immediately or a summary after several minutes;
writing down an unfinished topic so the client knows it has not been forgotten.
These strategies can reduce cognitive load for both people. They are most useful when agreed collaboratively rather than imposed as remedial social-skills training.
The client is seeking reassurance that does not last
Some clients repeatedly describe a situation because they want confirmation that they are not at fault, that another person still cares about them, or that a feared outcome will not occur. The counsellor may provide reassurance, but its effect fades quickly and the client begins assembling the case again.
Excessive reassurance-seeking is associated with depression and interpersonal rejection, although the relationships are complex and cannot be reduced to a simple causal sequence (Starr & Davila, 2008). In therapy, repeated reassurance may produce brief relief while leaving the client increasingly dependent on another person to resolve uncertainty.
The counsellor can validate the underlying fear without repeatedly deciding the question for the client:
“You want to know with certainty that you did nothing wrong. Each time I reassure you, it seems to help briefly, and then the doubt returns. I think we may need to understand what makes the reassurance disappear.”
This approach is different from withholding ordinary human reassurance. It becomes relevant when reassurance itself has become part of the maintaining cycle.
The speech represents a change in mental state
Talkativeness may occasionally require assessment rather than a conversational strategy. Rapid or pressured speech that is markedly different from the client’s baseline can occur alongside mania, hypomania, substance effects, acute anxiety or other changes in mental state.
Ask about the onset and context. Notice whether the speech is episodic, difficult to interrupt, increasingly tangential or accompanied by reduced need for sleep, elevated or irritable mood, unusual energy, impulsivity, racing thoughts, grandiosity or changes in functioning.
A lifelong tendency to think aloud is different from a sudden period of sleeping for three hours, beginning several major projects and speaking at a pace the client cannot regulate. Containment strategies should not substitute for appropriate assessment and referral.
Practical ways to contain extended talking
Make interruption part of the agreement
Counsellors often wait for a natural pause that never comes. By the time they interrupt, they are frustrated or trying to end the session. The interruption then feels sharper than it needed to be.
Discuss the possibility early:
“You often have several connected things you want to explain, and I want to help us keep track of the ones that matter most. I may sometimes pause you to summarise or ask us to stay with something. How would that be for you?”
The client may have useful preferences. Some people want an immediate signal when they have moved off topic. Others need a few minutes to complete their thinking before they can tolerate redirection.
Permission does not make every interruption helpful. Research from couple therapy suggests that therapist interruptions can affect alliance and may interact with gender and power dynamics (Oka et al., 2021). The purpose, timing and manner of the interruption still matter.
Interrupt to protect something important
An interruption usually lands better when the counsellor explains what they are protecting:
“I want to pause you because what you just said about being relieved when the appointment was cancelled seems important. I don’t want us to skip over it!”
This communicates active listening. The counsellor is not stopping the client because the story is boring but because the story deserves more attention.
Reflect before you redirect
When a client has repeatedly experienced interruption, a bare redirection can feel dismissive. Show that you have followed the account before changing direction by summarising:
“You tried to raise the issue carefully, he turned it into a criticism of you, and you ended up apologising. I can understand why you are still going over the conversation. Could we look at the point where you stopped saying what you actually thought?”
Accurate reflection may reduce the client’s need to provide further evidence. It also gives them an opportunity to correct your understanding before you move on.
Hold the branches of the conversation
Some clients continue talking because they fear that an unfinished subject will disappear. The counsellor can hold the list:
“You have mentioned the argument with your sister, the email from work and what happened when you could not sleep on Sunday. Which one most needs our attention today?”
If the client chooses one and then returns to another, refer back to the agreement:
“I’m going to hold the work email for later. We agreed to stay with what happened with your sister first.”
This is collaborative agenda-setting. Agenda-setting is often poorly implemented even in mental health consultations, but when it is done well it can clarify what both people expect from the available time (Frankel et al., 2013).
Use embodied co-regulation
One strategy I use when a client is speaking very quickly and their breathing seems high and shallow is to meet them close to their pace, but just beneath it. I then gradually slow my speech and breathing over several turns, sometimes repeating the sequence. I pay attention to the client’s response and check how the pace is feeling, rather than assuming that speech or breathing alone tells me exactly what is happening for them. Sensorimotor psychotherapy describes the use of embodied resources and therapist modelling to support regulation, although this particular pacing sequence is a practice-based strategy rather than a tested protocol (Lohrasbe & Ogden, 2017).
For a client who feels overwhelmed or disconnected from the present, orienting to the room can be one option. I might ask, “Would it help to look around the room together and name a few things you can see?” or invite them to notice the support of the chair, their feet on the floor, or an object they can hold. The client can choose what feels useful; I would not assume the room feels safe or insist on focusing inward. Grounding guidance includes orienting to current surroundings and using sensory anchors, and trauma practice guidelines recommend supporting present-orientation and grounding while monitoring the client’s level of activation (SAMHSA, 2014; American Psychological Association, 2024). If attention to the body or breath increases distress, I would offer another option or stop.
Use time openly
Time becomes easier to manage when it is discussed throughout the session rather than announced during the final minute.
Useful markers include:
“We are about halfway through. Do you want to continue with the story, or use the remaining time to think about what you might do?”
“We have 15 minutes left, and I want to leave enough time to consider what you need after today.”
“There are five minutes remaining. I am going to stop us there and summarise what I think we have identified.”
Regular time markers are particularly useful in brief work, EAP counselling and sessions where the client has difficulty estimating time.
Ask for the version that serves the work
Clients do not always know how much context the counsellor needs. You can make the request specific:
“What is the main thing you want me to understand about this?”
“Which part of the story connects most closely with what you want help with?”
This is usually more helpful than asking someone to “get to the point”, which assumes they already know what the point is and are choosing not to provide it.
Move between narrative, emotion and action
A session can become stuck when it remains entirely at one level. The counsellor can help the client move between:
what happened;
what the client thought and felt;
what the event means to them;
what pattern it may reflect;
what they want to do now.
For example:
“I think I understand the events. Could we move to what you made of them, what they meant for you?”
Or:
“We have spent some time understanding why this impacted you. Would it be helpful to discuss where to next?”
This movement should respond to the purpose of the session. Not every painful account needs to end with an action plan. It is still useful for the counsellor and client to know whether they are telling, understanding, experiencing, deciding or practising.
Talk about the pattern between you
When extended talking recurs, the interaction itself may need attention:
“I have noticed that I become very quiet in our sessions and wait for an opening. You keep talking, perhaps because I am not giving you a clear signal that I want to come in. I wonder what that is like from your side.”
Another possibility is:
“When I interrupt, you sometimes begin speaking faster and add more evidence. I wonder whether my interruption feels as though I have not heard you?”
Metacommunication allows the client and counsellor to examine the pattern together. It also creates an opportunity to repair any misattunement. Alliance repair is associated with better psychotherapy outcomes, although successful repair requires genuine attention to the client’s experience rather than a polished explanation from the therapist (Eubanks et al., 2018).
End the session when the session ends
Allowing sessions to run over can reinforce the idea that the client must keep talking until everything has been covered. It also creates practical problems for the counsellor and the next client.
A clear ending can be warm and firm:
“I need to stop us because we have reached the end of our appointment. I know this is unfinished. I have written down where we were, and we can decide next time whether to begin here.”
If the client introduces significant new material at the end, assess anything that requires immediate attention. Otherwise, acknowledge it and arrange how it will be addressed. Consistent endings are part of the treatment frame, not a failure of empathy.
What tends not to help
Waiting passively for the client to become less talkative rarely changes an established pattern. Neither does hinting through eye contact, shifting posture or repeatedly glancing at the clock.
It is also unhelpful to decide too quickly that the talking is resistance, avoidance, ADHD, autism or a personality characteristic. The same behaviour can have different functions across clients, and several functions may operate at once.
Repeatedly allowing the entire session to pass and then feeling resentful places responsibility on the client for managing a process they may not recognise. The counsellor is responsible for monitoring the time, remembering the agreed purpose and raising concerns about how the work is proceeding.
At the other extreme, rigidly enforcing an agenda can prevent important material from emerging. Therapist responsiveness involves adjusting interventions to the client and to what is happening moment by moment, rather than applying the same technique in every session (Esposito et al., 2024).
Bring the counsellor’s response to supervision
Clients who talk continuously can evoke strong responses. The counsellor may become bored, tired, irritated, helpless, cognitively overloaded or unusually passive. They may mentally leave the conversation and then feel guilty about it.
These reactions can offer information about how other people experience the client, but that is only one possibility. The response may also reflect the counsellor’s personal history, preferred conversational style, current fatigue, difficulty with conflict, or anxiety about exercising authority.
Research indicates that unexamined countertransference reactions are associated with poorer outcomes, while effective management of these responses is associated with better psychotherapy outcomes (Hayes et al., 2018). Supervision can help separate what belongs to the client, what belongs to the counsellor and what is being created between them.
Useful supervision questions include:
What happens in me while the client is talking?
At what point do I give up and stop trying to participate?
What do I imagine will happen if I interrupt?
Does the client respond differently when they feel accurately understood?
Is the talking moving the work forward, regulating distress or preventing something else from happening?
Have the client and I explicitly agreed on what they want from these sessions?
Am I calling my passivity “being person-centred” because taking more leadership feels uncomfortable?
How might culture, gender, disability, neurotype, age or professional authority affect how my interruptions are experienced?
What could I say in the next session that would make this process discussable?
Containment requires formulation
A lonely client, a client who thinks aloud, a client seeking accurate empathy, a client caught in rumination and a client entering a hypomanic episode may all speak for most of a session. The resemblance at the level of behaviour does not make the clinical task the same.
Effective containment begins with understanding the function of the talking. It may involve listening for longer, offering a more accurate reflection, making the structure explicit, interrupting earlier, slowing the pace, addressing avoidance, supporting conversational organisation or assessing a change in mental state.
The client remains central to the work, while the counsellor takes responsibility for the process. That includes protecting time for reflection, noticing when the conversation is circling, and speaking when silence from the counsellor is no longer serving the client.
Patterns like these are worth taking to supervision because they often contain much more information than “this client talks too much”. They tell us something about how the client regulates emotion, seeks understanding, organises experience and relates to other people. They may also tell us something useful about what the counsellor finds difficult to do.
We offer supervision! To learn more, click here.
References
Alderson-Day, B., & Fernyhough, C. (2015). Inner speech: Development, cognitive functions, phenomenology, and neurobiology. Psychological Bulletin, 141(5), 931–965. https://doi.org/10.1037/bul0000021
American Psychological Association. (2024). Guidelines on key considerations for working with adults with PTSD and traumatic stress disorders. https://www.apa.org/about/policy/adults-ptsd-traumatic-stress-guidelines.pdf
Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G., & Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704–1712. https://doi.org/10.1177/1362361320919286
Elliott, R., Bohart, A. C., Watson, J. C., & Murphy, D. (2018). Therapist empathy and client outcome: An updated meta-analysis. Psychotherapy, 55(4), 399–410. https://doi.org/10.1037/pst0000175
Esposito, G., Cuomo, F., Di Maro, A., & Passeggia, R. (2024). The assessment of therapist responsiveness in psychotherapy research: A systematic review. Research in Psychotherapy: Psychopathology, Process and Outcome, 27(2), Article 751. https://doi.org/10.4081/ripppo.2024.751
Eubanks, C. F., Muran, J. C., & Safran, J. D. (2018). Alliance rupture repair: A meta-analysis. Psychotherapy, 55(4), 508–519. https://doi.org/10.1037/pst0000185
Frankel, R. M., Salyers, M. P., Bonfils, K. A., Oles, S. K., & Matthias, M. S. (2013). Agenda setting in psychiatric consultations: An exploratory study. Psychiatric Rehabilitation Journal, 36(3), 195–201. https://doi.org/10.1037/prj0000004
Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40(2), 218–227. https://doi.org/10.1007/s12160-010-9210-8
Hayes, J. A., Gelso, C. J., Goldberg, S. B., & Kivlighan, D. M. (2018). Countertransference management and effective psychotherapy: Meta-analytic findings. Psychotherapy, 55(4), 496–507. https://doi.org/10.1037/pst0000189
Kessler, P. B., & Ikuta, T. (2023). Pragmatic deficits in attention-deficit/hyperactivity disorder: Systematic review and meta-analysis. Journal of Attention Disorders, 27(8), 812–821. https://doi.org/10.1177/10870547231161534
Kuo, J. R., Fitzpatrick, S., Ip, J., & Uliaszek, A. (2022). The who and what of validation: An experimental examination of validation and invalidation of specific emotions and the moderating effect of emotion dysregulation. Borderline Personality Disorder and Emotion Dysregulation, 9, Article 15. https://doi.org/10.1186/s40479-022-00185-x
Laurenceau, J.-P., Barrett, L. F., & Pietromonaco, P. R. (1998). Intimacy as an interpersonal process: The importance of self-disclosure, partner disclosure, and perceived partner responsiveness in interpersonal exchanges. Journal of Personality and Social Psychology, 74(5), 1238–1251. https://doi.org/10.1037/0022-3514.74.5.1238
Lohrasbe, R. S., & Ogden, P. (2017). Somatic resources: Sensorimotor psychotherapy approach to stabilising arousal in child and family treatment. Australian and New Zealand Journal of Family Therapy, 38(4), 573–581. https://doi.org/10.1002/anzf.1270
Milton, D. E. M. (2012). On the ontological status of autism: The “double empathy problem”. Disability & Society, 27(6), 883–887. https://doi.org/10.1080/09687599.2012.710008
Newman, M. G., & Llera, S. J. (2011). A novel theory of experiential avoidance in generalized anxiety disorder: A review and synthesis of research supporting a contrast avoidance model of worry. Clinical Psychology Review, 31(3), 371–382. https://doi.org/10.1016/j.cpr.2011.01.008
Oka, M., Seedall, R. B., Robinson, W. D., & Walker, J. (2021). Therapist interruptions: An examination of gender and therapeutic alliance in couple therapy. The American Journal of Family Therapy, 49(4), 409–425. https://doi.org/10.1080/01926187.2020.1823282
Rimé, B. (2009). Emotion elicits the social sharing of emotion: Theory and empirical review. Emotion Review, 1(1), 60–85. https://doi.org/10.1177/1754073908097189
Rose, A. J. (2002). Co-rumination in the friendships of girls and boys. Child Development, 73(6), 1830–1843. https://doi.org/10.1111/1467-8624.00509
Starr, L. R., & Davila, J. (2008). Excessive reassurance seeking, depression, and interpersonal rejection: A meta-analytic review. Journal of Abnormal Psychology, 117(4), 762–775. https://doi.org/10.1037/a0013866
Substance Abuse and Mental Health Services Administration. (2014). Trauma-informed care in behavioral health services (Treatment Improvement Protocol [TIP] Series 57; HHS Publication No. [SMA] 13-4801). https://www.ncbi.nlm.nih.gov/books/NBK207201/
Watkins, E. R., & Roberts, H. (2020). Reflecting on rumination: Consequences, causes, mechanisms and treatment of rumination. Behaviour Research and Therapy, 127, Article 103573. https://doi.org/10.1016/j.brat.2020.103573