SEEK: Salford Environment for Expertise and Knowledge

April 2014

Reliability of three objective hand measures in rheumatoid arthritis: joint circumference, composite finger flexion and the grip ability test

Hammond, A & Jones, V & Prior, Y 2014, 'Reliability of three objective hand measures in rheumatoid arthritis: joint circumference, composite finger flexion and the grip ability test', Rheumatology, 53(Suppl 1), p.i24.



Background: In preparation for a feasibility study evaluating compression gloves in rheumatoid arthritis (RA), we searched AMED (1985- April 2013) and MEDLINE (1988 – April 2013) to identify reliable methods of evaluating objective measures of hand joint swelling, finger motion and hand function. As data were being collected by therapists as part of their usual glove provision, measures needed to be brief, replicable in practice, and use easily obtainable, low-cost equipment. The feasibility study included multiple assessors, therefore we evaluated inter- and intra-rater reliability of measures.
Method: We selected the Figure of Eight method (hand swelling), composite finger flexion (CFF) to the distal palmar crease (DPC) (finger motion), and the Grip Ability Test (GAT,  hand function). During piloting (n=3), therapists had difficulty performing the Figure of Eight and CFF (DPC) methods and we changed to measuring metacarpophalangeal and proximal interphalangeal joints (MCPJs, PIPJs) circumference using a Jobskin® tape measure and CFF to the distal wrist crease (DWC). The GAT measures function bilaterally with the dominant hand performing most actions. As gloves can be prescribed for one hand only, we revised instructions to repeat the GAT for both left and right hands and further standardised instructions. A hand assessment manual was developed. Eleven therapists then completed reliability testing. Following training, they practiced the three measures with each other and two patient research partners for 1.5 hours, then were tested twice at one hour intervals on subjects’ right hands. Ten performed the GAT. Paired t-tests were used to identify differences between therapists’ test 1 and 2 (T1,T2) measurements (intra-rater reliability). Intra-class correlation coefficients were calculated (two-way random, consistency model) using ICC (2,11) across the 11 assessors (ICC(2,10), for the GAT) (inter-rater reliability).
Results: For all joint swelling measures, there were no significant differences between tests indicating good intra-rater reliability (eg 2nd-5th MCPJ swelling: T1 = 18.75 (SD 0.13); T2 = 18.77 (SD 0.14); p = 0.72). Inter-rater reliability (ICC (2,11)) varied from 0.19 (poor) for the wrist to 0.91 (almost perfect) for the 2nd PIPJ. For CFF, there were no significant differences between tests, indicating good intra-rater reliability (eg 2nd digit T1=9.38 (SD 0.31), T2=9.40 (SD 0.29); p = 0.74). Inter-rater reliability (ICC(2,11) ranged from 0.76 – 0.93 (i.e. strong to almost perfect). For the GAT, there were no significant differences indicating good intra-rater reliability (T1=24.66 (SD 4.24), T2= 24.59 (SD 4.20); p = 0.82).   Inter-rater reliability (ICC (2,10) was 0.98 (almost perfect).

Conclusion: To our knowledge this is the first reliability testing of joint circumference measured using Jobskin® tapes and CFF to the DWC in RA. Joint circumference, CFF and GAT measures had good intra- and mostly good inter-rater reliability


SEEK Members

External Authors

V Jones

Y Prior

Publication Details

Rheumatology, 53(Suppl 1), p.i24.