Measurement Properties of the Mayo-Portland Adaptability Inventory (MPAI-4) and Related Measures: A Systematic Review

Objective(s): (1) To assess the MPAI-4 and related measure’s (e.g., Participation Index (M2PI)) measurement properties and evidence quality; and (2) identify the interpretability and feasibility of these measures.

Data Sources: Two independent reviewers searched nine electronic databases and trials registries from inception to 2021 using the 2018 COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) guidelines for systematic reviews.
Study Selection: Forty-eight of the 605 retrieved peer-reviewed primary articles concerning the MPAI-4 or related measures were retained.

Data Extraction: We extracted study and patient characteristics, critically appraised risk-of-bias (COSMIN checklist) and rated the extracted results.

Data Synthesis: We used COSMIN’s approach to synthesize measurement properties (insufficient, sufficient), and modified GRADE approach to assess evidence quality (very low, low, moderate, high) by diagnosis (traumatic brain injury (TBI), stroke) and setting (inpatient, outpatient). The MPAI-4 is sufficiently comprehensible (GRADE: very low), with no evidence reported for its relevance or comprehensiveness. The MPAI-4 and M2PI have sufficient interrater reliability for stroke and TBI outpatients (GRADE: moderate), whereas interrater reliability between TBI inpatients and clinicians is insufficient (GRADE: moderate). There is no evidence for measurement error. The MPAI-4 and M2PI have sufficient construct validity and responsiveness across TBI inpatients (GRADE: high) and outpatients (GRADE: high), and stroke outpatients (GRADE: moderate-high). There is one study with mixed insufficient/sufficient evidence for each adapted version (21- and 22-item MPAI, 9-item M2PI) (GRADE: low-high).

Conclusions: In response to recent outcome standardization initiatives using the MPAI-4 and M2PI, this systematic review provides diagnosis and setting-specific supporting evidence for each measure. The MPAI-4 and M2PI have sufficient evidence to support their use in TBI inpatient and outpatient, and stroke outpatient populations. Adapted versions of the MPAI-4 warrant further study to generate clear recommendations for use. Since the MPAI-4 and related measures are formative as opposed to reflective models, structural validity, internal consistency and cross-cultural validity results are uninterpretable. Future studies should focus on measurement properties congruent with a formative model, especially reliability, measurement error, content validity and predictive validity.

Author(s) Disclosures: The authors have no conflicts of interest.