Table 2.
Minimal important change values for adult PROMIS physical function
| References | PROMIS measure | Language | Population | Intervention | Method used | Follow-up | Na | Correlation of PROMIS change score with anchor | MIC valueb | Comments |
|---|---|---|---|---|---|---|---|---|---|---|
| Bartlett et al. [78] | V1.0 Physical Function 20a | English (US) | Rheumatoid arthritis | Medication | Mean change of individuals responding a little better or a little worse on an anchor question on change in physical function | 4.6 (2.4) Months | 33 | Not reported | Improvement: 2.8 | |
| 46 | Deterioration: 1.4 | |||||||||
| 33 | Improvement: 2.5 | |||||||||
| V1.0 physical function 4a | 45 | Deterioration: 1.8 | ||||||||
| Bernstein et al. [47] | V1.0 physical function | English (US) | Carpal tunnel release | Surgery | Mean change in subgroup who reached a MIC on MHQ Function (23 point) or BCTQ (0.74 points) | 6 Weeks or 3 months | 52 (MHQ) | Not reported | 1.8 | MIC of MHQ Function was based on MHQ satisfaction, where ‘satisfied’ was defined based on effect size (distribution-based) |
| 40 (BCTQ) | 2.8 | |||||||||
| Hays et al. [51] | V1.0 physical function 19 (out of 20a) | English (US) | Rheumatoid arthritis | Any treatment | Mean change in patients who reported “a little better” or “a little worse” on an anchor question on how you are feeling now compared to 6 months ago |
0–6 Months 6–12 months |
35 | 0.33 |
Improvement: 1.6 2.0 |
Anchor question does not refer to change in physical function |
|
0–6 Months 6–12 months |
113 |
Deterioration: 0.8 1.7 |
||||||||
| Hung et al. [53] | V1.2 physical function CAT | English (US) | Foot and ankle disorders | Surgery | Optimal ROC cut-off point to distinguish patients who reported much worse, worse, improved and much improved from patients who reported slightly worse, no change, slightly improved on an anchor question on change in physical function | 3 Months |
32 vs 42 (43% changed) |
Not reported | 12.0 | Patients who were improved and patients who were deteriorated were lumped together in the analyses |
| > 3 Months |
219 vs 337 (39% changed) |
7.9 | ||||||||
| > 6 Months |
128 vs 178 (42% changed) |
10.5 | ||||||||
| Hung et al. [54] | V1.2 physical function CAT | English (US) | Pathology of the hip or knee | Surgical and non-surgical interventions | Optimal ROC cut-off point to distinguish patients who reported much worse, worse, improved and much improved from patients who reported slightly worse, no change, slightly improved on an anchor question on change in physical function | 3 Months |
54 vs 88 (38% improved) |
Not reported | 2.0 | |
| > 3 Months |
366 vs 577 (39% improved) |
3.4 | ||||||||
| 6 Months |
21 vs 34 (38% improved) |
3.5 | ||||||||
| > 6 Months |
192 vs 421 (31% improved) |
8.2 | ||||||||
| Katz et al. [70] | Physical Function 4a | English (US) | SLE | Not reported | Mean change of patients who reported somewhat better on an anchor question on change in functioning | 4 Periods of 6 months | 141 | Not reported | Improvement: 0.8 | Each patient was included up to 4 times in the analysis |
| 169 | Deterioration: 1.1 | |||||||||
| Kazmers et al. [86] | V2.0 physical function CAT | English (US) | Non-shoulder hand and upper extremity pathology | Recovering from surgery, undergoing surgery, corticosteroid injection, other | Mean difference between patients reporting no change and patients reporting (slightly) improved on an anchor question (2 anchor questions) | 6 (± 4) Weeks | 381 | Not reported | 2.1, 3.6 | |
| Kenney et al. [56] | Physical function CAT | English (US) | patients undergoing knee arthroscopy | knee arthroscopy | Optimal ROC cut-off point to distinguish patients who reached a MIC (11.5 points on 0–100 scale) on the IKDC | 2 Weeks to 12 months |
51 vs 25 (67% improved) |
0.76 | 3.3 | Version not reported |
| Khutok et al. [87] | Physical function 4a (part of PROMIS-29 v2.1) | Thai | Chronic low back pain | Many received standard physical therapy | Mean change of individuals reporting little improvement on an anchor question on change in pain interference | 4 Weeks | 54 | 0.26 | 0.1 | |
| Kuhns et al. [88] | Physical function CAT | English (US) | Patients undergoing hip arthroscopy, (including 61 athletes) | Hip arthroscopy | Optimal ROC cut-off point to distinguish patients who reported their level of function after surgery excellent or good versus patients who reported their level of function after surgery average, fair or poor | 21.3 (± 4.4) Months | 113 | Not reported | 5.1 | Anchor question about current (postoperative) level of functioning instead of change |
| Lapin et al. [59] | V1.0 physical function CAT | English (US) | Ischemic and hemorrhagic stroke patients | Routine care | Mean change in patients who indicated minimally or much improved/worse on an anchor on change in function | 5–6 Months | 167 | 0.28 | Improvement: 4.0 (± 7.4) | Minimally and much improved patients were combined, MIC likely overestimated |
| 34 | Deterioration: 0.3 (± 7.0) | |||||||||
| Lee et al. [60] | V1.0 physical function 10a | English (US) | Adults (40+) with knee OA | Tai Chi or physical therapy | Mean change in patients that reached 1–2 MICs on SF-36 PF | 12 Weeks | 13–28 | 0.59 | 1.9 to 2.2 |
Unclear which MIC values for SF-PF were used and whether they were anchor-based Lower bound of MICs was increased to > SEM |
| Lee et al. [76] | V1.2/V2.0 physical function CAT | English (US) | Thumb carpometacarpal arthritis | Operative or non-operative treatment | Mean change score in the mild improvement group according to an anchor question on change in your condition since last visit | 63 days (IQR 42–125) | 70 | Not reported | 3.9 (95% CI 3.3–4.7) | Anchor question does not refer to change in physical function |
| Sandvall et al. [64] | V1.2/V2.0 physical function CAT | English (US) | Unilateral distal radius fracture | Nonsurgical care | Mean change in patients who reported mildly better on an anchor question on change in your condition | Median 35 days (IQR, 25 − 45) | 138 | Not reported | 3.6 | Anchor question does not refer to change in physical function |
| Smit et al. [90] | 24-item PROMIS physical function geriatric rehabilitation (PROMISPF-GR) short form | Dutch (NL) | patients admitted for geriatric inpatient rehabilitation | Usual care in nursing homes | predictive modeling to distinguish patients who reported to be improved versus patients who reported to be unchanged on an anchor question on change in physical function | 41 Days | 167 | 0.32 | 8.0 (95% CI 4.1–12.5) | |
| Speck et al. [91] | V1.0 physical function 13-item custom short form | English (US, Canada, EU, and Australia) | Tenosynovial giant cell tumors | pexidartinib or placebo | Mean change in patients who changed 1 point on a 5-point global rating of physical function | 25 Weeks | 27 | Not reported | 4.0 | Design and sample size not clearly reported |
| Stephan et al. [99] | V2.0 physical function 4a | German (Switzerland) | Foot and ankle disorders | Orthopedic foot and ankle surgery | Optimal ROC cut-off point to distinguish patients who reported operation did help or operation helped a lot from patients who reported operation helped only a little, did not help or made things worse | 6 Months |
166 vs 36 (82% improved) |
0.45 | 4.6 |
Anchor does not refer to change in physical function MIC overestimated due to high percentage of patients improved |
| Yost 2011 [45] | V1.1 cancer physical function-10 | English (US) | Advanced-stage cancer |
Chemotherapy only (74.3%) Chemo- and radiation therapy (9.9%) Other mixed modalities (13.8%) Missing 2.0% |
Mean change in patients who changed 1–2 MICs on SF-36 PF (8–16 points) Mean change in patients who reported a little better or moderately better or a little worse or moderately worse |
6–12 Weeks | Subgroups of 88 | > 0.30 |
Median 3.0 (range 3.0–3.0) |
PROMIS Cancer scales are on the same metric as the PROMIS generic item banks Estimates for improvement and deterioration were lumped together MIC of SF-36 partly based on distribution-based methods |
BCTQ boston carpal tunnel questionnaire, CAT computerized adaptive testing, IKDC international knee documentation committee, IQR inter quartile range, MIC minimal important change, MHQ Michigan hand questionnaire, NA not applicable, NRS numerical rating scale, OA osteoarthritis, PROMIS patient-reported outcomes measurement information system, ROC receiver operating characteristics, SEM standard error of measurement, SF-36 PF short form 36, physical function subscale
aN reflects the number of patients on which the presented MIC values are based (often a subset of the study population)
bMIC values for minimal important improvement, unless otherwise specified. For all values, higher MIC values indicate more improvement or more deterioration for the construct being measured