Assessing Change in Psychoanalytic Psychotherapy of Children and Adolescents (Psychology, Psychoanalysis & Psychotherapy)

1. 1Establish the treatment alliance immediately after introductions: state plainly the length of treatment, the schedule of repeat assessments, confidentiality, and specifically what may be shared with parents at reviews. Set these boundaries before therapeutic work begins.
2. 2Name the one limit to confidentiality up front — if the child raises suicidal intent or self-harm putting them at risk, this may be shared with responsible adults. Do this so safety and trust are not later confused.
3. 3Convey the method: tell the child the work is developing an understanding of the meaning of all communications between you — verbal and non-verbal — and that missed sessions, reluctance to come, and reluctance to leave are all part of that meaning, held within the boundary of the therapy room.
4. 4Provide the secure base for exploration by tracking, session to session, the child's vitality and range of expressed emotion; treat increases in shown emotion as the early signature of a deepening alliance rather than as disruption.
5. 5Expect and tolerate the sequence: alliance-building first, then deepening, then the emergence of aggression, conflict, and working through — noting that working through often only becomes visible after six months to a year.
6. 6Test the strength of the bond by whether the child can hold onto and express aggressive or negative feelings toward you in the transference and still leave the room with the relationship intact.
7. 7Read the same change against the severity of the disorder: with a severely disturbed child, a deepening of alliance is a turning point; with a neurotic child the identical shift is only 'the process goes on.'
8. 8Link alliance quality to observable task engagement — check whether a stronger bond coincides with richer symbolic play or fuller narrative accounts of personal experience, since alliance should be associated with that engagement.

In Frida's therapy, the therapist described a turning-point session at the end of the first year (Session 57). Frida was able to hold onto aggressive and negative feelings toward the therapist in the transference across the whole session, and she left the room still carrying them. In the preceding session aggressive impulses had already broken through, so they had been 'on their way' — the material was building, not erupting at random. The therapist read this not as a rupture but as evidence that the alliance was secure enough to contain hostility without collapse. Because Frida was not severely disturbed, this deepening is closer to 'the process goes on' than to a dramatic turning point, illustrating how the same event is graded against the child's diagnosis. The sequence — early alliance-building, then deepening vitality and emotion, then aggression and working through around the one-year mark — tracks the expected process arc.

• If the child is severely disturbed, then treat any deepening of the alliance as a turning point; if the child is neurotic, then read the same deepening as ordinary progress.
• If aggressive or negative feelings surface in the transference and the child can hold them and still leave the room, then the alliance is strong enough to support working through — do not rush to repair.
• If material such as suicidal intent or self-harm emerges and the child is at risk of harm, then break confidentiality and share with responsible adults, as flagged at the outset.
• If a stronger bond coincides with richer symbolic play or fuller personal narratives, then use that engagement as your outcome-linked marker of alliance quality.
• If conflict and working through have not appeared by six months to a year, then re-examine whether the alliance actually deepened past the initial building phase.

• Deferring boundary-setting (length, reviews, confidentiality limits) until after work begins — the tell is confusion or a sense of betrayal when parents are informed at a review.
• Reading emerging aggression in the transference as alliance breakdown and prematurely soothing it — the tell is you defuse hostility instead of letting the child leave the room still holding it.
• Grading a change without reference to the child's diagnosis, so a routine shift gets over-celebrated or a genuine turning point gets missed — the tell is your significance judgment doesn't move with severity.
• Treating only verbal content as the communication — the tell is you ignore missed sessions, reluctance to come, and reluctance to leave as meaningful data.
• Expecting working through too early and judging the therapy stalled — the tell is impatience before the six-month-to-year window in which conflict typically becomes visible.

• ▸Phase now: alliance-building / deepening / conflict & working through
• ▸Boundaries set: length, review schedule, confidentiality limits (yes/no + date)
• ▸Child severity: severely disturbed vs neurotic (sets turning-point threshold)
• ▸Vitality & emotion: what more of the child's aliveness/feeling appeared this session
• ▸Transference test: could the child hold negative/aggressive feeling and still leave the room?
• ▸Non-verbal communications noted: missed sessions, reluctance to come/leave
• ▸Task engagement: change in symbolic play or personal narrative this session
• ▸Change grade: turning point / turning-point-like / process goes on (given severity)
• ▸Months in treatment: is this within/after the 6–12 month working-through window?

1. 1Establish a play typology before you code anything: distinguish representational/thematic play (child creates narrative structures for affect-laden relationships) from non-representational play (repetitive board games) — per Target et al. (2005), not all play in therapy is equal.
2. 2Score the child's play along the four structural components of the Children's Play Therapy Instrument (Kernberg, Chazan & Normandin, 1998): affective, cognitive, dynamic, developmental, plus the quality of the relationship within the play. Foreground the cognitive component — the child's social representations and capacity to create narrative structures to represent different affect-laden relationships.
3. 3Track range and depth as a change measure across sessions, not a one-off: the hypothesis you are testing is that positive change on the cognitive/representational dimension predicts better outcome, so you need repeated ratings on the same construct.
4. 4Match technique to symbolic capacity. For children with sufficient symbolic capacity, use tactfully timed, dosed and phrased transference interpretation. For children with insufficient symbolic capacity (traumatised, abused, psychotic/borderline) prepare the ground first using description/mirroring before any interpretation.
5. 5Use descriptive commentary on play sequences to signal the child is being attended to and thought about, and simple reflection of the events and feelings they describe, as the technique that builds the alliance and scaffolds symbolic engagement.
6. 6Decide your scaffolding stance explicitly: if the goal is helping the child with the rudiments of narrative, choose whether you guide by asking about setting, temporal ordering and character development, or whether you merely reflect the story line — and note which you did, since which is more change-conducive is an open question.
7. 7For maltreated children age 4–8, elicit generic child–parent representations using standardised story stems (MacArthur Story Stem Battery / Story Stem Assessment Profile), read the play narrative as how the child reflects upon reality rather than as literal fact, and rate presence/absence of themes to compare against later placement or treatment change.
8. 8Separate acting-out/enactment from symbolic material: when events concretely represent distress rather than being symbolised (e.g. self-harm reported misleadingly), review risk and cross-check in team meetings rather than treating it as play communication.

Hodges and colleagues devised the Story Stem Assessment Profile (formerly the 'Little Pig' stems) specifically to explore the inner representational world of children aged 4–8 who have been subjected to maltreatment or neglect. They built it from the MacArthur Story Stem Battery, adding five new stems to probe maltreatment issues. Sessions are videotaped, transcribed, and rated for the presence or absence of thirty-two themes, each scored on a three-point scale, with trained raters consistently achieving good reliability. The working assumption is that children's representations build up over time from expectable interactions, so the stems aim to elicit generic child–parent representations most likely to affect later relationships. Critically, the child's play narrative is treated as an indicator not of reality but of how 'the child reflects upon reality.' Hodges et al. then compared these representations before and after adoptive placement, using the same rated dimensions to detect change — a concrete model for treating symbolic material as both medium and measurable agent of change.

• If the play is repetitive and non-representational (e.g. a board game), then do not score it as symbolic engagement — it lacks the representational quality of thematic play.
• If the child has insufficient symbolic capacity to sustain transference work, then prepare the ground with description/mirroring first and dose interpretation tactfully, rather than delivering standard verbal interpretations.
• If an interpretation risks the child hearing 'you are to blame for everything that went wrong,' then reframe to acknowledge actual external circumstances and confine the reference to the present feeling toward the therapist.
• If the material is enactment that concretely represents distress rather than symbolised communication, then trigger a risk review and verify in team meetings rather than interpreting it as narrative.
• If the child's cognitive/representational play scores rise across sessions, then predict and check for corresponding gains in mentalization and emotion regulation, since improved symbolic play is hypothesised to work through those mechanisms.

• Counting play frequency or interpretation dose instead of assessing quality — the tell is you can report how much play/interpretation happened but nothing about its accuracy or representational depth.
• Treating all play as equivalent, so repetitive games get logged as symbolic progress — the tell is high 'play' scores with no narrative structure or affect-laden relationship content.
• Delivering a clumsily phrased transference interpretation to a low-symbolic-capacity child — the tell is escalating anger or withdrawal, and the child conflating the therapist with the family situation.
• Reading the child's play narrative as literal reporting of events rather than as how the child reflects upon reality — the tell is you start fact-checking the story instead of tracking its representational shifts.
• Missing acting-out disguised as narrative (e.g. self-harm reported as if long ago) so risk goes unassessed — the tell is that the inaccuracy only surfaces later in team meetings.
• Over-structuring the child's narrative with leading questions when reflection might drive more change, or under-scaffolding a child who needs the rudiments of narrative — the tell is you never decided or recorded which stance you took.

• ▸Play type: mark representational/thematic vs non-representational (board game/repetitive).
• ▸Affective component: note range and intensity of affect expressed in play.
• ▸Cognitive component: rate capacity to create narrative structures for affect-laden relationships.
• ▸Dynamic component: note conflicts/defences enacted in the play.
• ▸Developmental level: rate the maturity of the play organisation.
• ▸Relationship quality in play: describe the child–therapist relation within the activity.
• ▸Therapist stance used: record description/mirroring, reflection, structuring questions, or interpretation.
• ▸Range/depth change vs prior session: mark increased / stable / decreased with a one-line reason.
• ▸Enactment/risk flag: note any concrete (non-symbolic) representation of distress requiring team review.

1. 1Distinguish interpretive from facilitative activity before intervening: facilitative moves (description, mirroring, holding projections) sustain the symbolic process; interpretive moves name meaning and push the child toward more verbal-level reflection. Default to facilitation until the symbolic material is stable enough to bear interpretation.
2. 2Read play as symbolic representation of concerns, conflicts, fears, and urges that exceed the child's capacity for verbal representation or provoke anxiety requiring indirect expression. Track the range and depth of the play, not just its content, because increases in range and depth are what predict better outcomes.
3. 3Facilitate expression first: help the child bring thematic material into play, then assist them in understanding the meaning and relevance of the play productions for everyday emotional life. Do not skip the facilitation step to reach an interpretation faster.
4. 4Dose interpretation to what the child can reconsider. Following Erikson, children who transfer the unsolvability of their problems into play and onto the observer need to be induced by systematic interpretation to reconsider, on a more verbal level, the constellations that have overwhelmed them — meaning interpretation is timed to move symbolic content toward verbal reflection, not to display therapist insight.
5. 5Prioritize meaningful over frequent interpretation. Mere frequency of interpretations does not predict better outcomes; make interpretations that facilitate insight and expansion of self-awareness, and expect these to be more strongly associated with change than other verbalizations such as requests for information.
6. 6Protect the alliance as the medium for symbolic engagement. The quality of the child-therapist relationship is central and plausibly works by supporting the child's involvement in therapeutic tasks — the expression of symbolic material — so read a drop in symbolic play as a possible alliance signal, not just resistance.
7. 7Watch for transformation of core relationship themes across sessions. Treat the child's thematic play (and adolescent relationship narratives) as data on interpersonal representations, and time deeper interpretation to points where these representations are shifting.

A child repeatedly stages a scene where a small figure is left unsolvable and unrescued, and directs the therapist to fail at helping — transferring the unsolvability of the problem into the play and onto the observer. You begin facilitatively: you describe what you see ('the little one keeps getting left, and no one can fix it') and hold the projected helplessness rather than reacting to it, which keeps the symbolic process going and lets the range of the play widen. Over sessions the play deepens and the theme recurs with variation, signaling a workable interpersonal representation. Only then do you interpret systematically, inducing the child to reconsider on a more verbal level — linking the play's constellation to how it feels when things at home seem unfixable. You do not multiply interpretations; you make one meaningful link tied to expanding self-awareness. The tell of success is not compliance but increased depth and range in the subsequent play and the child's own verbal reflection on it.

• If the child's material exceeds verbal capacity or triggers enough anxiety to need indirect expression, then stay facilitative (describe, mirror, hold) and let play carry the theme before attempting interpretation.
• If symbolic play is expanding in range and depth, then continue and consider timed interpretation; if it narrows or collapses, then treat it as a possible alliance rupture and return to facilitation.
• If you are about to interpret mainly because the meaning is obvious to you, then stop — frequency and therapist certainty do not predict outcome; interpret only when it will help the child reconsider on a more verbal level.
• If the child is transferring unsolvability onto you as observer, then hold the projection and use systematic (not one-off) interpretation to move the overwhelming constellation toward verbal reflection.
• If a recurring relationship theme is beginning to transform across sessions, then time your deeper interpretive work to that shift rather than to a fixed schedule.

• Interpreting to prove insight rather than to expand the child's self-awareness. Tell: you feel clever after the intervention but the play does not deepen or the child disengages.
• Equating more interpretations with more progress. Tell: high interpretive frequency logged, but outcome and symbolic depth flat — you are treating interpretation like a dose count.
• Collapsing the facilitation step and jumping straight to naming meaning. Tell: the child had no chance to express the theme in play first, and the interpretation lands as intrusion.
• Mistaking narrowing play for resistance rather than an alliance signal. Tell: you push harder interpretively while the relationship (and the symbolic process) is actually fraying.
• Failing to hold projected helplessness — reacting to being made to fail as observer instead of containing it. Tell: you become directive or reassuring, and the unsolvable constellation never moves toward verbal reconsideration.

• ▸Session play theme: one line naming the concern/conflict/fear/urge represented
• ▸Range & depth this session: rate wider/same/narrower vs last session
• ▸Facilitative moves used: describe / mirror / hold projection — which and when
• ▸Projection held: what helplessness or role was placed on you, and how you contained it
• ▸Interpretation made: verbatim link offered (leave blank if none warranted)
• ▸Timing rationale: why now — what made the material ready for verbal reconsideration
• ▸Child response: shift in verbal reflection or in subsequent play depth
• ▸Alliance signal: any drop in symbolic engagement to revisit next session

1. 1Establish a stable, accepting relational context before targeting mentalization; the therapeutic alliance is the precondition for the child producing symbolic material at all. Treat alliance quality as a distinct process target, not incidental.
2. 2With younger children, facilitate the expression of symbolic material through play; your task is to broaden self-experience by increasing the range, depth, and emotional richness of play rather than delivering interpretations as the primary change lever.
3. 3With older children and adolescents, shift the medium to talk, but hold the same aim: broaden and deepen self-experience to enhance reflective function.
4. 4Time and titrate interpretations rather than maximizing their frequency; aim interpretations at facilitating insight and expanding self-awareness, and prefer meaningful interpretations over neutral verbalizations like requests for information.
5. 5Track the child's capacity to interpret their own and others' behavior in terms of internal mental states, since behavior becomes meaningful and predictable through inferences about mental states.
6. 6Monitor movement from raw affect toward mentalized affect, on the assumption that mentalization in play develops structures for regulating emotion; watch for accompanying gains in emotion regulation capacity.
7. 7Separate two process types in your notes: the treatment procedure (your facilitation of symbolic play/expression) versus the treatment target (the child's mentalizing capacity as the pathogenic mechanism), so you can tell whether your intervention is actually moving the target.

Fonagy and Target (2002) identify 'the ability to focus attention as needed' and 'the ability to behave as appropriate' as characteristic problems of AD/HD children, and note both functions carry important meaning for the development of reflective capacities. Working within this frame, a therapist treating an AD/HD child does not begin by explaining the child's behavior to them. Instead, they create an accepting relationship in which the child's play can widen in range and emotional depth, so that impulsive or dysregulated acts start to be represented symbolically rather than only enacted. As the child's play begins to hold and name internal states, the deficits in self-regulatory skills that Carney (2002) places 'at the heart' of the syndrome become workable targets. Understanding and trying to modify these deficits, per the passages, is a shared goal across psychoanalytic schools. The predicted sequence is: therapist facilitation raises symbolic play, richer play raises mentalization, and increased mentalization builds structures that improve emotion regulation.

• If the child cannot yet produce symbolic material, then work on alliance and safety first, because the alliance drives engagement in symbolic play and other symbolic productions.
• If play range and depth are increasing over sessions, then expect and look for corresponding gains in mentalizing and, downstream, in emotion regulation; if play broadens but regulation does not, revisit whether mentalization is actually deepening.
• If you are counting interpretations to gauge progress, then stop—mere frequency of interpretations is unlikely to predict better outcomes; weigh whether each interpretation is meaningful and expands self-awareness.
• If the patient is an older child or adolescent, then use talk as the medium but keep the target identical to play work: broadening and deepening self-experience to enhance reflective function.
• If the child's problems are 'mixed' rather than 'pure,' expect a harder or slower response, consistent with better outcomes for pure emotional disorders in short focused treatment.

• Treating interpretation as the primary change mechanism; the tell is a session log dominated by therapist explanations while the child's play stays narrow and affectively flat.
• Collapsing procedure into target—assuming that because you facilitated symbolic play, mentalization improved; the tell is claimed progress with no independent evidence of expanded self/other mental-state inference.
• Rushing interpretations without timing and titration or a stable relational context; the tell is the child withdrawing, resisting, or ceasing symbolic production after your comments.
• Using attachment security as a proxy for improvement; Gerber's caution shows that surface security may mask work still needed, so a 'secure-looking' presentation can be misread as change.
• Expecting immediate outcome differences and abandoning the approach early; the Muratori 'sleeper' effect shows gains may appear only at later follow-up, so premature judgment discards effective work.

• ▸Alliance quality (this session): note stability of relational context and child's engagement
• ▸Symbolic material produced: play content, verbalizations, or personal narrative expressed
• ▸Play range and depth rating: has range/emotional richness broadened vs. prior session (down/same/up)
• ▸Mental-state references: instances child named own or others' internal states
• ▸Interpretation log: which interpretations were meaningful and expanded self-awareness (not a count)
• ▸Timing/titration note: was interpretation delivered within a stable context; child response
• ▸Emotion regulation signs: evidence of new structures for regulating affect emerging
• ▸Procedure vs. target: what YOU did (facilitation) vs. change in the child's mentalizing capacity
• ▸Pure vs. mixed presentation: note to calibrate expected pace and possible sleeper-effect timing

1. 1Distinguish 'mild' from 'peak' affect states as you work: the material accessible through mild-activation states supports normal internalization and can be connected to images and words, while peak affects (terror, rage, euphoria, the desire for fusion) overstrain the child's capacity for stimulus processing and arrive as embodied, split 'psychic truths' rather than verbal reports.
2. 2Attend to non-verbal channels as the primary route to non-conscious emotion. Because much communication takes place non-verbally, treat your counter-transference experience as data about what the child cannot yet feel or say, and make active use of it rather than setting it aside.
3. 3Work the negative transference deliberately rather than avoiding it. Track the negative feelings aroused by termination — issues of identity and sexual identity often surface at the end — and stay with them, since good outcome is associated with the therapist's ability to deal with the negative transference.
4. 4Anchor the affective work in the child's real support system: engage parents/carers in parallel work (in the depression trial, one parent session per two child sessions) since external support and parental willingness to engage predict better progress.
5. 5Use fortnightly supervision to contain and process the material. The intensity the child evacuates lands on you; supervision safeguards against 'wild' interpretations and counteracts the temptation to avoid the negative transference by reminding you that facing it predicts good outcome.
6. 6Sequence expectations honestly: some change is immediate and some is a 'sleeper' effect that only appears at follow-up, so continue working through emotion even when surface measures look flat, and don't read early plateau as failure.
7. 7Hold to the treatment frame around missed sessions: count sessions missed for reasons to do with acting out (child's or caregiver's), treating the enactment itself as processable emotional material rather than mere absence.

In the Trowell multi-centre depression trial, children aged 10-14 all meeting criteria for major depressive disorder and/or dysthymia received focused individual psychodynamic psychotherapy — a mean of 24.7 fifty-minute sessions — with a separate case worker running one parent session for every two child sessions. Therapists worked the emotional material in-session while supervision contained the counter-transference load. At end of treatment 74.3% of the individual-therapy cases were no longer clinically depressed, comparable to family therapy; but the affective work kept consolidating after termination. By six-month follow-up 100% of the individual-therapy cases were no longer clinically depressed, versus 81% for family therapy, and individual therapy proved effective across major depression, dysthymia and 'double depression' with no relapses. The persistent, ongoing improvement shows that working through emotion in session builds regulation structures that hold and extend beyond the treatment window.

• If the child's presentation is dominated by internalizing/emotional problems, then expect strong responsiveness to in-session emotional work; if externalizing dominates, then expect harder going unless the externalizing is coupled with internalizing problems or treatment frequency is increased.
• If the child communicates mainly non-verbally, then treat your counter-transference as the working material and take it to supervision before interpreting.
• If fundamental issues (identity, sexual identity, loss) surface near termination, then address the negative transference directly rather than softening or extending treatment to avoid it.
• If a session is missed for reasons connected to acting out by child or caregiver, then count and work it as meaningful emotional communication.
• If surface outcome measures show no change at six months, then continue the work — a sleeper effect may produce significant gains only at two-year follow-up.

• Avoiding the negative transference, especially around termination — the tell is a therapist trained in long-term work who feels 'cruel' stopping and drifts into reassurance instead of naming the negative feelings; good outcome depends on facing them.
• Interpreting from counter-transference without supervisory containment — the tell is 'wild', unsubstantiated interpretations that outrun the evidence; fortnightly supervision exists precisely to prevent this.
• Reading peak-affect material as literal report — the tell is taking split idealized/persecutory 'psychic truths' at face value rather than as embodied memories that overstrained the child's processing capacity.
• Declaring failure at end-of-treatment when measures look flat — the tell is stopping engagement early and missing the sleeper effect that only registers at longer follow-up.
• Working the child in isolation from carers — the tell is neglected parent/carer work and unstable placement; progress tracks with stable placement, external support and parental willingness to engage.

• ▸Session number / frame status: note attendance and whether any absence links to acting out (child or carer).
• ▸Dominant affect register today: mild-activation (workable, verbalizable) vs peak affect (terror/rage/euphoria/fusion).
• ▸Non-verbal communications observed: what the child enacted or evoked rather than said.
• ▸Counter-transference note: what you felt, and what it may signal about non-conscious emotion.
• ▸Negative transference present? What was named or avoided, especially termination-related.
• ▸Interpretation offered vs held: grounded in evidence or flagged for supervision.
• ▸Parent/carer work this cycle: session done, engagement quality, placement/support status.
• ▸Supervision item to raise: containment need or interpretation to check.
• ▸Change signal: immediate shift, plateau, or possible sleeper effect to revisit at follow-up.

1. 1Define the target outcome domain before selecting measures: decide whether the primary distal outcome is psychic (mood, anxiety), social-communicative (dyadic/triadic relating, attachment), somatic (physical markers like diabetic control, growth), or global functioning — the domain dictates the instrument.
2. 2Select measures matched to the population and to the change you expect to see. When you doubt conventional psychological outcome measures capture the change (as Moran et al. did), choose physical primary outcomes — glycosylated haemoglobin, height velocity — instead of forcing symptom questionnaires.
3. 3Assess at three fixed time points at minimum: intake/pre-treatment, termination/end of therapy, and follow-up (six months as in Trowell 2007, or one year as in Moran, or eighteen-month intervals as in Gerber). Use an independent assessor at these points.
4. 4Set thresholds for 'clinically significant' before you start: use diagnostic status change (no longer meets MDD/dysthymia criteria per Kiddie-SADS), movement into an 'acceptable' clinical range (HbA1c), or a reliable-change criterion — not just mean group shifts.
5. 5Track diagnosis rates and comorbidity together, since reduction in the primary disorder plus reduction in comorbid conditions is the fuller functioning outcome (Trowell reported both).
6. 6Extend follow-up to detect a sleeper effect: continue measuring after termination because response can keep improving post-completion (Individual Therapy rose to 100% no longer depressed at six-month follow-up).
7. 7Interpret proxy measures cautiously: do not treat attachment security as equivalent to symptomatic improvement — in Gerber, successful cases moved from secure toward preoccupied-entangled before resolving, contradicting the proxy assumption.
8. 8Report constraints alongside results: name randomisation status, sample size, dropout, and cases lost to follow-up that inflate or deflate response rates.

Trowell et al. (2007) treated children meeting criteria for major depressive disorder and/or dysthymia, assessed with the Kiddie-SADS and Child Depression Inventory. They compared 8-14 sessions of Family Therapy (mean 11) against 16-30 sessions of Individual Therapy (mean 24.7) plus parent sessions, assessing pre-treatment, at end of therapy, and at six-month follow-up. At treatment end, 74.3% were no longer clinically depressed after Individual Psychotherapy and 75.7% after Family Therapy, with an overall reduction in comorbid conditions. Gains persisted with no relapses in the six months after treatment, and by follow-up 100% (Individual) and 81% (Family) were no longer clinically depressed. The continued improvement after completion is read as a 'sleeper effect' — an ongoing response to therapy. The apparent 20% Individual advantage was partly an artefact of four Family Therapy cases lost to follow-up, showing why dropout must be reported when interpreting response rates.

• If conventional psychological measures seem unlikely to capture the change you expect, then select physical/functional primary outcomes (e.g. HbA1c, height velocity) and treat symptom scales as secondary.
• If treating disruptive/conduct disorder, then expect harder engagement and higher dropout, and consider intensive treatment over three years — under those conditions the outcome gap versus emotional disorders disappears (Fonagy & Target).
• If a case appears securely attached at intake yet has severe character pathology, then do not read early security as improvement; expect a necessary-but-not-sufficient shift toward preoccupied-entangled before a return to secure signals structural change.
• If response rates differ between groups at follow-up, then check for cases lost to follow-up before crediting the difference to treatment.
• If depression is mild-to-moderate and the child is 10-14, then favour a talking therapy first — medication has not been shown effective in this age group, and watchful waiting then psychological therapy is the recommended sequence.

• Equating attachment security with recovery — the tell: you score AAI 'secure' at intake and conclude no work is needed, missing that improvers first moved to preoccupied-entangled (Gerber).
• Reporting only end-of-treatment figures and stopping — the tell: you miss the sleeper effect and understate outcomes that keep improving to follow-up (Trowell's rise to 100%).
• Letting dropouts inflate an apparent advantage — the tell: a 20% between-group gap that shrinks once four lost-to-follow-up cases are accounted for.
• Using generic symptom questionnaires for conditions where they don't capture the change — the tell: null results in cases (diabetes, autism) where physical markers or dyadic-relating ratings would have shown clear improvement.
• Overclaiming from quasi-randomized or retrospective small samples — the tell: a strong efficacy statement (10 of 12 improvers in psychoanalysis) without flagging small n, sequential assignment, and confounders like fewer Axis II subjects in one arm.

• ▸Primary outcome domain: state psychic / social-communicative / somatic / global functioning
• ▸Primary measure + rationale: name instrument and why it captures your expected change
• ▸Clinical-significance threshold: define diagnostic-status change or reliable-change / 'acceptable range' cutoff in advance
• ▸Assessment time points: list intake, termination, and follow-up interval (e.g. 6 or 12 months)
• ▸Assessor independence: who rates, and are they blind to condition
• ▸Comorbidity tracking: how you record change in secondary/comorbid conditions
• ▸Follow-up / sleeper-effect check: how you detect continued post-treatment improvement
• ▸Attrition log: record dropouts and cases lost to follow-up with reasons
• ▸Design constraints statement: randomisation status, sample size, confounders to disclose

1. 1Distinguish structural from symptomatic change before you measure anything: symptom reduction on aggregate outcome measures is one axis; attachment security, ego-flexibility, self-regulation and reflective capacity are the structural axis. Track them separately.
2. 2Administer the Adult Attachment Interview (or age-appropriate narrative equivalents like Hodges & Steele's story-stem assessments) at initial assessment, and again at termination and beyond, at matched time points, so you can chart movement rather than a single snapshot.
3. 3Read a mid-treatment shift toward preoccupied-entangled classification in an initially 'secure' patient with severe character pathology as possible destabilisation of a defensive surface security — a necessary but not sufficient step — not as deterioration.
4. 4Look for a return-to-secure classification at or beyond termination as the marker of consolidated structural change; time your final assessment to catch it, since it may only appear after treatment ends.
5. 5Assess across multiple domains and methods (multi-method, multi-informant): individual psychodynamic ratings, family-functioning measures, and developmental status, because change in one domain can coincide with apparent deterioration in another.
6. 6Build in follow-up assessment (one-year, four-year) to detect sleeper effects and forestalled future difficulties that immediate post-treatment measures miss.
7. 7Combine narrative process description (the meaning-making, story-telling account from case material) with empirical instruments; neither alone captures structural change.

Gerber (2004) compared psychoanalysis and psychodynamic psychotherapy in patients with severe character pathology, using symptomatic outcome and the AAI. Ten of the 12 symptomatic improvers were in the psychoanalysis group. Strikingly, 54% were classified secure on the AAI at initial assessment — yet in the successful treatments, patients moved toward a preoccupied-entangled pattern that then began to resolve by termination. Gerber reads the initial 'security' in these patients as defensive rather than genuine, so the shift into preoccupied-entangled is interpreted as necessary movement, with a return to secure at or beyond termination taken as evidence of true structural change. He is careful to flag the quasi-randomised design, small sample, and confounds such as fewer Axis II subjects in the psychoanalysis group, and recommends replication with the AAI at matched time points in a larger sample.

• If an initially 'secure' patient with severe character pathology shifts to preoccupied-entangled mid-treatment, then treat it as expected destabilisation, not failure — and continue toward the expected resolution.
• If you only have a single AAI at intake, then do not use attachment security as a proxy for therapeutic improvement — you cannot distinguish genuine from defensive security without repeated measures.
• If a group improves on individual measures but shows deterioration on family-functioning measures (as in Szapocznik's individual-therapy group without parallel parent work), then check whether the wider system was engaged before attributing the change to the individual treatment.
• If immediate post-treatment measures are flat, then schedule follow-up (one to four years) before concluding no structural change occurred — sleeper effects and harnessed family resilience may emerge later.
• If sample is small, heterogeneous, and manual adherence unchecked, then present structural findings as hypothesis-generating and recommend replication at matched time points rather than generalising.

• Equating attachment security with therapeutic improvement — the tell: a patient rated 'secure' at intake with no repeated AAI, so defensive security is mistaken for structural health.
• Reading a mid-treatment move to preoccupied-entangled as regression and pulling back — the tell: destabilisation interpreted as harm rather than as a necessary-but-not-sufficient step.
• Measuring only symptom aggregates and declaring structural change — the tell: outcome reported without any attachment, reflective-capacity or developmental-status instrument.
• Ignoring the family/system context, running individual work without parallel parent work — the tell: individual gains alongside deteriorating family-functioning scores at follow-up.
• Stopping assessment at termination — the tell: no follow-up window, so sleeper effects and forestalled future difficulties go undetected; also failing to check manual adherence or run intention-to-treat with variable dropout, biasing conclusions.

• ▸Patient character pathology level: note severity and any Axis II features
• ▸Domain — attachment: AAI/story-stem classification at intake
• ▸Domain — attachment: classification at termination and post-termination
• ▸Symptomatic axis: aggregate outcome measure score, intake vs. termination
• ▸Ego/self-regulation & reflective capacity: instrument used and change noted
• ▸Family/system domain: family-functioning score and whether parallel parent work occurred
• ▸Assessment time points: list matched intervals (intake, termination, 1yr, 4yr)
• ▸Method check: multi-method and multi-informant sources used
• ▸Design caveats: sample size, heterogeneity, manual adherence, ITT status

1. 1Name the third task explicitly. Before introducing research, state that the centre already holds two established tasks — psychoanalytic psychotherapeutic treatment and clinically oriented training — and that research is being added as a third, distinct culture that must be bridged, not merged by force.
2. 2Survey and describe before you collect. Produce a written description of the clinical work, survey the literature on child psychotherapy, and develop an overview of the institute's clinical work as a whole, so research questions grow out of practice rather than an a priori program.
3. 3Map staff attitudes and the force field. Systematically identify staff feelings about making research part of their work, then categorize facilitating and hindering factors by area of concern (identity, confidentiality, time, prior failures, career-versus-patient beliefs).
4. 4Change the organisation of daily work. Set aside protected time for discussing research ideas and build supervision or consultation about research projects into the clinic exactly as supervision is already a permanent feature of clinical work.
5. 5Design data collection that satisfies two audiences. Collect both quantitative and qualitative data on what treatment is given to which patients and how treatments work for different categories, standardized rigorously enough to make sense to outsiders while respecting contextual contingencies and individual variation — as quality assurance, not method validation.
6. 6Have leadership contain the surfaced conflicts. Once the structure changes, latent psychotherapist-versus-researcher conflicts become manifest; the head and staff group jointly listen to and metabolise anger, frustration, idealisation and hope, returning them in less fragmented form.
7. 7Value every contribution regardless of role or enthusiasm. Explicitly acknowledge within the staff group that some will formulate questions and analyse data while others interview or judge narratives, and treat differing levels of engagement as understandable and acceptable.
8. 8Sustain attention from the top. Recognise that without continuous attention from the head, systematic data collection diminishes; without engaged staff, data quality and quantity fail — so protect the mutual trust between staff and head over years, not months.

At the Erica Foundation in Stockholm — a centre providing child and adolescent psychotherapy plus university-level training — Siv Boalt Boëthius introduced research as a third task to a clinically oriented staff group. The strategy was empirical research based on clinical practice and modern psychoanalytic theory: systematically collecting quantitative and qualitative data to show what treatment is given to which patients and how treatments work across patient categories. Staff surfaced real resistance — that research might undermine their identity as clinicians, take over the 'real work' with patients, and repeat a prior project that generated much data but few publications of clinical interest. Leadership treated these reactions as material to be contained rather than overridden, and daily work was reorganised to set aside time for research discussion and consultation alongside existing clinical supervision. Over several years and many projects the general attitude toward clinically based research changed radically, with clearly increased awareness of the need for systematic documentation. Competition for recognition and space increased, which put continuous pressure on leaders to provide containment — and that containment yielded higher creativity that appears to have raised the quality of the clinical work itself.

• If the organisational structure has not yet changed, then expect the psychotherapist-versus-researcher conflict to stay latent — introduce research only when leadership is ready to contain the conflict it will make manifest.
• If a staff member believes research serves the researcher's career while therapy serves the patient, then reframe the task as the same stance of questioning and uncertainty the clinician already uses in therapy, rather than debating the belief.
• If continuous attention from the head lapses, then assume systematic data collection will diminish — treat sustained leadership engagement as a non-negotiable precondition, not a nicety.
• If the aim is to test how the method is implemented in this institution, then use quality-assurance evaluation and standardisation as a data-collection guide — do not import evidence-based-medicine RCT logic to validate the method itself.
• If a staff member contributes only interviews or independent judgements rather than leading projects, then acknowledge that role as acceptable and value it equally, because uneven engagement is normal and forcing uniformity destroys creativity.

• Collecting impressively but publishing nothing useful. The tell: staff reference a prior project that 'involved several persons but generated few publications' of limited clinical interest — fear of repetition signals this pattern is already feared.
• Merging the two cultures instead of linking them. The tell: boundaries between clinical and research work dissolve and clinicians feel research is taking over 'the real work,' rather than each area keeping its boundary within a shared organisational space.
• Leaving leadership containment implicit. The tell: rising competition for recognition and space goes unmetabolised, surfacing as anger and fragmentation rather than being digested and returned in acceptable form.
• Bolting research onto unchanged daily work. The tell: no protected time for research discussion and no research supervision exist, so the third task competes with clinical hours and quietly starves.
• Ignoring confidentiality mechanics. The tell: no agreed method for writing about a patient without identification, so clinicians resist data sharing on ethical grounds and collection stalls.

• ▸Existing tasks named — list the current core tasks (e.g., treatment, training) the new task must bridge.
• ▸Clinical work description — one-line summary of what is being done, for whom, drawn from literature survey and institute overview.
• ▸Research question grounded in practice — the 'which treatment, which patients, what outcome' question this project answers.
• ▸Facilitating factors by area — enablers listed under identity, time, leadership, confidentiality.
• ▸Hindering factors by area — resistances (identity threat, prior failure, career belief) listed under the same areas.
• ▸Daily-work changes required — the protected time and research-supervision slots to be added to the schedule.
• ▸Data plan — quantitative and qualitative data to collect and the standardisation rule that keeps it credible to outsiders.
• ▸Confidentiality method — how a patient will be written about without being identifiable outside the therapy.
• ▸Containment owner and cadence — who metabolises staff reactions, how often, with final responsibility named to the head.

1. 1Restrict use to clinicians who already hold a completed psychoanalytic child training, have undergone personal therapy, and are in supervision — the manual does not teach psychoanalytic psychotherapy and is no substitute for clinical training.
2. 2Open the manual with the disorder frame: introduce adolescent depression, then set out the psychodynamic framework for understanding depression in young people (e.g. depressive-position anxieties, fears about aggression, failure of containment) so technique rests on an explicit rationale.
3. 3Lay out the treatment steps the therapist follows in sequence, covering crisis intervention, coexisting marital and family conflict, transitions, and — flagged as most important — termination.
4. 4Write explicit management guidance for serious clinical situations: the suicidal young person, violence, substance abuse and self-harm, stating which tasks belong to the therapist during the work.
5. 5Close with technical issues specific to brief, time-limited work, including how to handle negative transference and the negative feelings aroused by termination when fundamental identity issues surface at the end.
6. 6Wrap the manual in supervision: schedule fortnightly supervision to process and contain distress, use counter-transference material, and guard against unsubstantiated 'wild' interpretations.
7. 7Build in adherence checking — the passages repeatedly note that adherence to the manual was not checked, which is a stated limitation; specify how (e.g. audiotape monitoring) adherence will be verified before treatment begins.
8. 8Set an operational rule set even for administrative events, e.g. count a missed session against treatment if it is missed for reasons of acting out by the child or caregiver, but not if the family was unaware of the meaning of their behaviour.

In the Trowell, Rhode and Hall chapter, a treatment manual was written for depression in late childhood and early adolescence and used within a multicentre trial (Tsiantis et al., 2005; Trowell et al., 2007) comparing individual psychodynamic therapy with family therapy. The manual moved from an introduction to adolescent depression, into the psychodynamic framework, then into ordered treatment steps and management of crises, marital/family conflict, transitions and termination, ending with technique for brief work. It was delivered by trained therapists and supervisors, with fortnightly supervision built in because of the severity of distress and the heavy reliance on non-verbal, counter-transferential communication. The authors state plainly the manual does not train people to do psychoanalytic psychotherapy — only already-trained, personally-analysed, supervised clinicians can use it. Notably, the reviewers observe the authors did not discuss adherence to the manual across the different centres, and a related Heidelberg study explicitly reported that adherence to the manual was not checked. This shows the construct working (a shared, rationale-driven document producing an empirically testable treatment) while exposing its recurring weak point: unverified adherence.

• If a would-be user has not completed a psychoanalytic child-psychotherapy training, had personal therapy, and secured supervision, then withhold the manual — it is not a training substitute.
• If the young person presents suicidality, violence, substance abuse or self-harm, then apply the manual's serious-clinical-situation guidelines and assign the specified therapist tasks rather than improvising.
• If a session is missed, then count it against treatment only when the absence reflects acting out by child or caregiver; if the family was unaware of the meaning of their behaviour, do not count it.
• If the study aims to claim treatment fidelity, then check adherence (e.g. audiotape monitoring) — do not report outcomes as manual-driven without it.
• If fundamental issues such as identity or sexual identity surface near termination, then address the negative transference directly in supervision rather than softening or avoiding the ending.

• Treating the manual as a training tool — the tell is untrained or unsupervised staff using it and producing 'wild', unsubstantiated interpretations the manual explicitly warns against.
• Skipping the rationale and jumping to technique — the tell is treatment steps applied without an articulated psychodynamic understanding of depression, so choices look arbitrary.
• Never verifying adherence — the tell (seen in both the depression manual chapter and the Heidelberg study) is a published limitation that adherence 'was not checked', undermining any claim the treatment matched the manual.
• Under-resourcing supervision — the tell is therapists overwhelmed by non-verbal, counter-transferential distress with no fortnightly containment, drifting from the protocol.
• Avoiding termination work — the tell is therapists trained for long-term treatment failing to address the negative transference and the cruelty/tantalising feelings aroused by stopping time-limited work at a critical moment.

• ▸Clinician credential: confirm completed psychoanalytic child training, personal therapy, current supervision
• ▸Disorder frame recorded: depression rationale and psychodynamic framework documented before technique
• ▸Treatment-step sequence: crisis, marital/family conflict, transitions, termination mapped for this case
• ▸Serious-situation protocol: therapist tasks for suicidality/violence/substance abuse/self-harm specified
• ▸Brief-work technique noted: time-limit handling and negative-transference plan for termination
• ▸Supervision schedule: fortnightly sessions booked, counter-transference use planned
• ▸Missed-session rule applied: acting-out vs. unaware-of-meaning distinction logged per absence
• ▸Adherence monitoring method: audiotape/adherence check named and scheduled (do not leave blank)
• ▸Limitation note: sample/context caveats and any unchecked-adherence risk flagged for the write-up

1. 1Match dosage to diagnosis, not to a default: before setting frequency and length, ask which pattern of response the presenting problem shows — some difficulties respond to brief, minimal-contact work while others need longer, more frequent treatment. The optimal value varies by disorder and developmental stage.
2. 2When comparing or planning against another modality, control for the fact that different therapies peak at different points. Do not impose a fixed session count as if all treatments improve on the same curve — dynamic therapy improves proportionately more in early sessions, then more slowly (Howard et al., 1986).
3. 3For adolescents, consider framing a session-limited contract (e.g. thirty sessions) explicitly as an extended therapeutic assessment, using the limit to force work on separation and management of aggression rather than treating the cap as a failure of dose.
4. 4Always pair individual child work with parent/family work when family functioning is a target: individual psychotherapy alone tends to produce poor outcomes for the family, whereas parent work alongside individual work has demonstrated impact on family functioning.
5. 5Hold the setting constant as the container for intensity: same day, same time, same therapist, same room, at regular (e.g. once-weekly) intervals, so deep-lying fears of abandonment and destructiveness can be worked with predictably.
6. 6Plan and measure a follow-up window, not just end-of-treatment: 'sleeper effects' and slower, more sustained gains from individual work mean the comparison of intensities must extend to later follow-up (e.g. four years).
7. 7Use the family's own coping capacity as a variable in the dose decision — for less disadvantaged, resourceful families, a minimal-contact intervention may harness resilience and outperform longer contact.

In the Smyrios & Kirby (1993) randomized trial, children aged 5–9 with 'disturbances of emotion specific to childhood' were assigned to three arms of psychoanalytically informed family and individual treatment differing only in dose: time-limited, time-unlimited, and a four-session minimal-contact control. Stringent exclusion criteria (no single-parent families, no history of mental illness) meant the sample was relatively resourced. At four-year follow-up all three groups did well, but the group seen least did rather better. The researchers speculated the four-session control was most effective because the families' own capacities for coping and resilience had been harnessed rather than displaced. The operational lesson: more sessions is not automatically more benefit; for a resourced family a brief intervention can be the correct dose, and the advantage only becomes visible at long follow-up, not at termination.

• If the family is relatively resourced and low in disadvantage, then bias toward a briefer or minimal-contact contract and let their own resilience carry the change.
• If family functioning is itself a treatment target, then never offer individual child work alone — add parent work in conjunction, or expect poor family outcomes.
• If you are comparing your dose to another modality, then control for duration and expect different peak times — do not read an early plateau in one therapy as inferiority.
• If the patient is a young adolescent who fears entrapment, then a limited contract (used as extended assessment) can be more engaging than an open-ended long-term commitment — provided future access to therapy is realistic.
• If immediate outcome at termination is good, then expect durability into adulthood; if treatment ended unsuccessfully, expect a preoccupied/entangled attachment pattern to persist, so judge dose adequacy at termination.

• Prescribing a fixed number of sessions as a universal standard — the tell: you justify it by 'controlling' across therapies, ignoring that therapies with earlier peaks are penalized by the same count.
• Assuming longer and more frequent is always better — the tell: you are surprised when the briefest arm outperforms at follow-up, as in the randomized trial cited by Kennedy.
• Treating a session cap as a cruel imposition and rushing or avoiding the ending — the tell: you skirt the negative transference instead of using the limit to work on separation and aggression.
• Offering individual child psychotherapy without parent work when the family is impaired — the tell: the child changes but family functioning does not, producing poor outcomes for the family.
• Judging effectiveness only at termination — the tell: you miss sleeper effects and the slower, more sustained gains of individual work that only appear at later follow-up.

• ▸Presenting problem / diagnosis: name the difficulty and note its known response pattern (fast vs slow-but-sustained).
• ▸Developmental stage: child / young adolescent / young adult — flag entrapment fears and separation issues.
• ▸Family resource level: assess disadvantage and coping capacity; high resource favors briefer dose.
• ▸Frequency chosen: once-weekly vs intensive multiple sessions — state rationale.
• ▸Length / contract: open-ended vs session-limited (e.g. 30) — if limited, frame as extended assessment.
• ▸Parent/family work: yes/no and how combined with individual work (required if family functioning is a target).
• ▸Setting constants: same day, time, therapist, room confirmed.
• ▸Follow-up measurement point: set a later window (e.g. 4 years) to capture sleeper effects.
• ▸Termination outcome check: rate immediate outcome and note attachment implications for durability.

1. 1Ground competence in the four assessable inputs the passages name: formal training/certification, clinical experience, personal analysis, and supervision. Treat each as a distinct evidence stream you can document, not a vague self-assessment.
2. 2Secure personal experience of an actual psychoanalysis before treating young patients. Widlöcher's argument is that this is a precondition, not a bonus — it gives you the sensitivity and capacity to associate that you then pass on to the patient.
3. 3Make detailed process recording notes from memory (not from tape) after every session, and hand these to the Project Co-ordinator. These notes are the raw material for supervision and the primary window on how you handle transference and counter-transference.
4. 4Audio tape every session for the explicit purpose of assessing therapist competence and adherence to the manual — separate this recording function from your memory-based process notes so each does its intended job.
5. 5Bring the process notes into supervision, giving particular attention to the first and final sessions and the sessions around holiday breaks, where separation and counter-transference pressure are highest.
6. 6Distinguish, per Widlöcher, two evaluation questions and know which you are answering: (1) whether the method fits the pathology, versus (2) how well you as a particular practitioner implement it. Competence assessment is the second, more delicate question.
7. 7Preserve flexibility and tact as part of competence — the evaluation must take account of the therapist's capability to respond in an attuned, case-specific way, not just manual adherence.

In the child study design, a therapist treating a subject aged ten to fourteen works largely verbally, so an audio tape of each session serves the research purpose of rating therapist competence and adherence to the manual. Separately, the therapist writes detailed process recording notes from memory (not transcribed from the tape) and hands them to the Project Co-ordinator. In supervision these notes are examined, with particular scrutiny given to the first session, the final session, and sessions bracketing holiday breaks. Across the treatment the therapist builds a 'narrative' to help the child make sense of loss, trauma, separation and key relationships, working in a child-led order. Over the last ten sessions attention shifts to termination — reworking earlier topics and preparing for separation from the therapist. Competence here is visible in how the therapist reads body language for transference and counter-transference cues and handles the ending, not only in whether the manual was followed.

• If the practitioner has not had personal experience of an actual psychoanalysis, then question their capacity to pass on their own sensitivity and associative capability to the patient — competence is not only know-how.
• If you are evaluating a therapist, then decide first whether you are testing the method against the pathology or testing this practitioner's implementation; the second question demands finer, case-specific instruments.
• If the assessment relies only on manual adherence scores, then add ratings of flexibility and tact, because competence and capability are partly expressed through attuned deviation.
• If a session falls at the start, the end, or around a holiday break, then give it heightened supervisory attention — these are where counter-transference and separation pressures concentrate.
• If a practitioner resists research on grounds that quantitative judgment threatens their freedom, then use group work among practitioners to develop critical sense without erasing individual responsibility.

• Making process notes from the audio tape rather than from memory — the tell is verbatim, transcript-like notes that lose the therapist's own selective, associative processing that supervision needs.
• Treating personal analysis as optional credentialing — the tell is a therapist who can describe technique fluently but cannot access or use their own counter-transference with the young patient.
• Collapsing 'the method works' with 'this therapist works' — the tell is defending poor implementation by citing evidence for the approach in general.
• Reducing competence to manual adherence and ignoring flexibility and tact — the tell is high adherence scores alongside a rigid, unattuned relationship the child does not engage.
• Neglecting endings and break sessions in supervision — the tell is a termination phase that arrives abruptly, with no reworking of earlier topics or preparation for separation.

• ▸Formal training / certification: name qualification and issuing institute.
• ▸Clinical experience: years and populations (children/adolescents) worked with.
• ▸Personal analysis: state whether completed and by whom (precondition for competence).
• ▸Supervision arrangement: name supervisor and frequency; note who receives process notes.
• ▸Recording method: confirm every session audio taped AND process notes made from memory.
• ▸Adherence check: rating of fidelity to the manual for the rated week.
• ▸Flexibility/tact rating: brief note on attuned, case-specific responses this period.
• ▸Counter-transference use: one observation from body language or the relationship, with how it was handled.
• ▸Endings focus: for first/final/holiday-break sessions, note what termination or separation work was done.

1. 1Run parent work in parallel with the child's individual therapy from the outset — not as an afterthought. The passages show that individual psychotherapy without work with parents or family produces poor outcomes for the family, whereas parent work in conjunction with individual work impacts family functioning.
2. 2Assign a distinct parental counselor alongside the child's therapist so each has a protected role. In the Erica Foundation model (Carlberg, 2009), goal-directed, time-limited child psychotherapy runs with parallel parental counselling as separate but coordinated tracks.
3. 3Track parental states of mind explicitly, because they interplay with the child's representations of family life — this interplay is detectable as early as three months following an adoptive placement. Treat the parent's internal picture as a therapeutic target, not context.
4. 4Decide the combination and sequencing of treatments per case: ask 'what is the best combination of treatments and in what order for any given clinical situation?' rather than defaulting to child-only work.
5. 5Gather concordant data from multiple informants — therapist, parental counselor, supervisor, child, and where relevant the special educational teacher — to identify whether change is occurring across the family system, not just in session.
6. 6Watch for the 'sleeper effect': individual work appears slower yet possibly more sustained, so plan follow-up rather than judging outcome at the point of ending. Keep focus on process and short-term goals, not only the expected end result.

In the three-centre childhood depression study (Trowell et al., 2007; Tsiantis et al., 2005), conducted in London, Athens and Helsinki, time-limited individual psychodynamic psychotherapy was compared with system-integrative family therapy — building parent and family work into the design rather than treating the child alone. The interplay between parental states of mind and the child's representations was shown to be observable as early as three months after adoptive placement, demonstrating why the family system is worked in parallel. The finding that individual work is slower yet possibly more sustained, while family/parent work impacts family functioning, drove the clinical question of which combination and order fits a given case. In the Erica Foundation study, Carlberg ran goal-directed, time-limited child psychotherapy with parallel parental counselling, using systematically collected narratives from patients, parents and therapists. One former child patient's question — whether there is 'too much emphasis on the individual' to solve the problem — captures precisely the risk this construct addresses. The concordance across informants (therapist, parental counselor, supervisor, child, teacher) about whether change had occurred showed the parallel tracks could be read together, even when they disagreed on timing.

• If individual child therapy is proceeding without any parent or family work, then expect poor outcomes for the family and add a parallel parent track.
• If the child's representations of family life are shifting, then examine the parents' states of mind in parallel, since the two interact within months.
• If outcome looks weak at the point of ending, then delay judgment — a sleeper effect means individual work is slower but possibly more sustained.
• If informants disagree on when change occurred but agree that it did, then treat the change as real and investigate its content and underlying factors rather than resolving the timing dispute.
• If a former patient signals 'too much emphasis on the individual', then treat that as evidence the family/system component was under-weighted.

• Treating the child in isolation: the tell is measurable child improvement alongside unchanged or worsening family functioning.
• Bolting on parent work late instead of running it in parallel — the tell is that parental counselling is reactive to crises rather than tracked from the outset.
• Ignoring parental states of mind as a target and treating parents only as reporters — the tell is that the child's representational shifts are recorded but the parent's internal picture is never assessed.
• Judging success only by the end result, missing the sleeper effect — the tell is discharging on the assumption that slow progress means failure.
• Collapsing the parental-counselor and child-therapist roles into one person, losing the coordinated dual perspective — the tell is no independent informant concordance to read change across the system.

• ▸Child therapist / Parental counselor: name each role holder separately
• ▸Child's presenting representations of family life: current picture, dated
• ▸Parental state(s) of mind: what internal picture each parent holds now
• ▸Projections to reclaim: what the parent attributes to the child that belongs to the parent
• ▸Interaction pattern maintaining disorder: the specific loop to change
• ▸Treatment combination and order: what runs when, and why for this case
• ▸Informant concordance log: therapist / parent counselor / supervisor / child / teacher — change yes/no + when
• ▸Sleeper-effect follow-up: date to reassess after ending, given slower-but-sustained pattern