Below are five precise, testable research questions that operationalize “level of arthritis” in older adults using validated patient-reported outcomes (PROs), objective performance measures, imaging, and biomarkers. Each includes outcomes, instruments, design, and analysis in brief.
-
Multidimensional Severity Index (MSI) validity in knee/hip osteoarthritis (OA)
Primary outcome: Construct validity (factor structure) and criterion validity (AUC for severe OA) of an MSI combining pain/function PROs, performance, and imaging.
Key instruments: WOMAC or KOOS/HOOS; 4‑meter gait speed; 30‑s chair-stand; 6‑minute walk distance (6MWD); radiographic Kellgren–Lawrence (KL) and/or MRI WORMS; handgrip strength as frailty anchor.
Design/population: Cross-sectional + 12‑month longitudinal cohort, n≥300 adults ≥70 with suspected or established OA.
Analysis: Exploratory/confirmatory factor analysis to derive MSI; ROC vs KL≥3 and MRI-based cartilage loss; responsiveness (SRM) over 12 months.
-
Hand arthritis severity: ultrasound-pathology vs functional capacity
Primary outcome: Correlation between ultrasound synovitis/osteophyte burden (grey scale + power Doppler summed score) and functional limitation.
Key instruments: Ultrasound GS/PD scoring of MCP/PIP/DIP; AUSCAN (pain/stiffness/function); grip and pinch dynamometry; Purdue Pegboard/Dexterity tasks.
Design/population: Cross-sectional study, n≥200 adults ≥70 with symptomatic hand OA/RA admixture; blinded US readers.
Analysis: Multivariable regression adjusting for age, sex, neuropathy, and sarcopenia; partial R² for US measures; minimal detectable change at 6 months.
-
Wearable-derived activity signatures vs disease activity in elderly rheumatoid arthritis (RA)
Primary outcome: Strength of association between daily gait speed/step-count variability and RA activity scores (DAS28, SDAI, RAPID3).
Key instruments: Wrist or hip accelerometer (24/7), morning-stiffness eDiary, DAS28/SDAI, RAPID3; CRP/ESR as inflammatory anchors.
Design/population: 6‑month prospective study, n≥150 adults ≥70 with RA; monthly clinic visits + continuous wearables.
Analysis: Mixed-effects models with random intercepts; Granger-style temporal precedence to test if wearable metrics anticipate flares by ≥7 days.
-
Biomechanics vs symptoms: IMU-estimated knee adduction moment (KAM) and pain
Primary outcome: Variance in WOMAC pain explained by IMU-inferred KAM compared with radiographic KL grade.
Key instruments: Thigh/shank inertial sensors + anthropometrics to estimate KAM; WOMAC pain; KL grade; body mass index; footwear/bracing records.
Design/population: Observational study (baseline) + 3‑month pragmatic brace/footwear optimization sub-study, n≥200 adults ≥70 with knee OA.
Analysis: Hierarchical models (KAM → pain) vs (KL → pain); Δpain after bracing stratified by baseline KAM tertiles.
-
Biomarker panels predicting structural progression in elderly OA
Primary outcome: Predictive accuracy of serum/urine biomarkers for 24‑month structural progression (MRI cartilage thickness loss or radiographic joint-space narrowing).
Key instruments: Serum hs‑CRP, IL‑6, COMP, MMP‑3; urine CTX‑II; MRI quantitative cartilage mapping or serial fixed‑flexion radiographs; pain/function PROs.
Design/population: 24‑month longitudinal cohort, n≥400 adults ≥70 with knee/hip OA; imaging at 0/24 months; biomarkers at 0/12/24 months.
Analysis: Penalized regression or random forest with repeated cross‑validation; AUC for “fast progressor” classification; net reclassification improvement over PRO+imaging only.