Three Sides You

Learn / Skills across approaches

Notice and repair a strain

The complete text is available below. Search, saving resources, and partner-practice tools require JavaScript.

Invite the person’s experience of a difficult moment and acknowledge your part before proposing a repair.

What it might sound like

“I think I missed something important. How did my response land?”

A common slip

Defending your intention instead of listening to the effect.

Across approaches

The purpose and timing depend on the formulation, relationship, and model. Follow a guide for its particular use rather than treating the wording as a universal script.

Sources and review scope

  • Book, clinical text, Paid · Purchase or library

    Negotiating the Therapeutic Alliance: A Relational Treatment Guide (opens in a new tab)

    Jeremy D. Safran and J. Christopher Muran (2000). Guilford Press

    The foundational account of alliance ruptures, therapeutic metacommunication, and stage-process models of repair. Its chapters move from rethinking the alliance to rupture resolution, brief relational therapy, and training.

    alliance, rupture and repair, metacommunication, relational

    Purchase or library access.

  • Review, process research, Free

    Repairing alliance ruptures. (opens in a new tab)

    Safran, J. D., Muran, J. C., & Eubanks-Carter, C. (2011). Psychotherapy, 48(1), 80-87. https://doi.org/10.1037/a0022140 (opens in a new tab)

    An early meta-analysis on rupture repair with clear practice recommendations, including the caution that frequent transference interpretation can harm and that the quality of an interpretation matters more than how often it is used.

    rupture and repair, alliance, training

    Free PDF hosted by a university clinic.

    Evidence appraisal
    Question
    Is repairing alliance ruptures associated with better outcome, and does rupture-resolution training help?
    Population
    3 studies (148 patients) for repair–outcome; 8 training/supervision studies (376 patients) for pre–post change; 7 studies (343 patients) with control conditions for the between-groups training comparison.
    Intervention
    Rupture repair; rupture-resolution training and/or supervision for therapists.
    Comparator
    Repair–outcome analyses were correlational. Training controls varied: waitlist, active treatments, supervised active treatments, and therapists serving as their own controls with different patients before and after training.
    Design
    Small meta-analyses. Training analyses separated within-group pre–post gains from between-groups contrasts with control conditions.
    Outcomes
    Treatment outcome.
    Follow-up
    Varies.
    Key results
    Repair–outcome r = .24 (95% CI .09 to .39), 3 studies, 148 patients. Training pre–post r = .65 (.46 to .78), 8 studies, 376 patients; this does not estimate improvement over a control. Training versus control r = .15 (.04 to .26), 7 studies, 343 patients. Excluding one outlier reduced the controlled estimate to r = .11 (.01 to .21), 6 studies, 321 patients.
    Limitations
    Very few small studies. Repair–outcome associations do not establish causation. The pre–post estimate cannot isolate training from therapy, passage of time, or other confounds. controlled designs and comparison conditions varied, and the estimate was sensitive to an outlier.