Assessing ad-hoc adaptations’ alignment with therapeutic goals: a qualitative study of lay counselor-delivered family therapy in Eldoret, Kenya
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
BACKGROUND: A key question in implementation science is how to balance adaptation and fidelity in translating interventions to new settings. There is growing consensus regarding the importance of planned adaptations to deliver interventions in contextually sensitive ways. However, less research has examined ad-hoc adaptations, or those that occur spontaneously in the course of intervention delivery. A key question is whether ad-hoc adaptations ultimately contribute to or detract from intervention goals. This study aimed to (a) identify ad-hoc adaptations made during delivery of a family therapy intervention and (b) assess whether they promoted or interrupted intervention goals. METHODS: Tuko Pamoja (Swahili: "We are Together") is an evidence-informed family therapy intervention aiming to improve family dynamics and mental health in Kenya. Tuko Pamoja employs a task-shifting model, delivered by lay counselors who are afforded a degree of flexibility in presenting content and in practices they use in sessions. We used transcripts of therapy sessions with 14 families to examine ad-hoc adaptations used by counselors. We first identified and characterized ad-hoc adaptations through a team-based code development, coding, and code description process. Then, we evaluated to what extent ad-hoc adaptations promoted the principles and strategies of the intervention ("TP-promoting"), disrupted them ("TP-interrupting"), or neither ("TP-neutral"). To do this, we first established inter-coder agreement on application of these categories with verification by the intervention developer. Then, coders categorized ad-hoc adaptation text segments as TP-promoting, TP-interrupting, or TP-neutral. RESULTS: Ad-hoc adaptations were frequent and included (in decreasing order): incorporation of religious content, exemplars/role models, community dynamics and resources, self-disclosure, and metaphors/proverbs. Ad-hoc adaptations were largely TP-promoting (49%) or neutral (39%), but practices were TP-interrupting 12% of the time. TP-interrupting practices most often occurred within religious content and exemplars/role models, which were also the most common practices overall. CONCLUSION: Extra attention is needed during planned adaptation, training, and supervision to promote intervention-aligned use of common ad-hoc adaptation practices. Discussing them in trainings can provide guidance for lay providers on how best to incorporate ad-hoc adaptations during delivery. Future research should evaluate whether well-aligned ad-hoc adaptations improve therapeutic outcomes. TRIAL REGISTRATION: Pilot trial registered at clinicaltrials.gov (C0058).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.013 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.008 |
| Science and technology studies | 0.003 | 0.001 |
| Scholarly communication | 0.000 | 0.002 |
| Open science | 0.002 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it