O-021 A New Domain Structure for the IMPACT-III, a Pediatric Inflammatory Bowel Disease (IBD) Health Reported Quality of Life (HRQOL) Tool
Bibliographic record
Abstract
The IMPACT-III is a validated, disease-specific HRQOL questionnaire for IBD patients ages 9 to 17. Items within IMPACT are grouped by domains, which cover different aspects of HRQOL. The domain structure has not been changed since validation, although 2 alternate domain structures have been proposed using different methodologies. Using a large, diverse cohort, the aim of this study was to develop the optimal domain structure for IMPACT. Three prospectively collected datasets of adolescents with IBD were used: (1) an internet-based cohort of children with IBD from the Crohn's and Colitis Foundation of America (CCFA) Partners Kids & Teens study, (2) the REACH infliximab trial for pediatric Crohn's Disease (CD), and (3) the T72 infliximab trial in pediatric Ulcerative Colitis (UC). Model development used a random sample of 75% of the data. Exploratory factor analyses were run in SPSS using Principal Axis Factoring. Confirmatory Factor Analysis compared previously developed and new factor models, using the full dataset. The Comparative Fit Index (CFI) determined model fit. Data from the CCFA dataset examined the psychometric properties of the new domain structure. Cronbach's alphas were calculated for reliability, while Pearson correlations measured construct validity between new IMPACT domains and PROMIS measures of anxiety, depression, fatigue, pain and peer relationships. The Pediatric Ulcerative Colitis Activity Index (PUCAI) and Pediatric Crohn's Disease activity index (PCDAI) defined disease activity. The sample included 502 patients, 125 with UC, 377 with CD (44% female, mean age = 13.4±2.4 years, 37% female, mean age of diagnosis = 10.6 ± 3.0 years). 56% were in remission, 13% had mild disease, and 28% had moderate/severe disease (3% missing). The original IMPACT and the Abdovic et al. domain structures had poor model fit (CFI = 0.60 and 0.70), while the Perrin et al. IMPACT-II model had reasonable model fit (CFI = 0.84). The new 4-factor IMPACT-III model was parsimonious and had good fit (CFI = 0.90). The new domains were: (1) General well-being (daily functioning, and “internal” symptoms such as stomach pain); (2) Social Functioning (social well-being, and “external” disease symptoms such as fecal urgency, which have a direct social impact); (3) Emotional functioning; (4) Body Image. The new domains and total scale had high internal consistency (Cronbach's alphas 0.92, 0.85, 0.86, 0.75, respectively; total scale 0.95). General well-being was most highly correlated with PROMIS pain and fatigue scales (Pearson's both −0.74), Social was highly related to anxiety and depression scales (both −0.64), Emotional was also highly related to the anxiety and depression scales (−0.54, −0.60), while Body Image was similarly related to all PROMIS instruments (0.44 for Peer Relations to −0.61 for depression). All correlations were highly significant (P < 0.001). The new domain structure of the IMPACT-III questionnaire exhibited reliable and valid domains, demonstrated better model fit and higher reliability than the original IMPACT domain structure, and showed construct validity with PROMIS instruments. These results have important applications both clinically and in large-scale clinical trials, particularly with interventions, which may have differential effects on specific aspects of HRQOL.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.014 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".