Background
Rheumatoid arthritis (RA) is a progressive, chronic, inflammatory, autoimmune disorder, affecting 0.5-1.5% of the world's population. Current RA disease activity assessment methods rely on subjective, labor-intensive methods. Therefore, an unmet need exists for objectively assessing RA disease activity through a low-labor method. It is hypothesized that a correlation exists between passive MCP joint stiffness and RA disease activity.

Objective
The objective of this study is to design, develop, and evaluate a measurement tool that can assess passive MCP joint stiffness of the index-, middle-, and ring fingers of both hands objectively. The tool will be evaluated on healthy participants.

Design methods
A structured approach is adopted for the design of the measurement tool. The design challenges are mapped out by writing down the requirements into four categories. In the next design phase, the partial functions and their corresponding potential solutions are worked out. Three concepts are developed which combine different sets of potential solutions. The most promising concept is determined through a multicriteria analysis and worked out into a final design. Before the evaluation process commences, the final design is reflected on and improved where necessary.

Evaluation methods
Firstly, the tool is verified by subjecting it to a linear torsion spring rather than a human MCP joint. Secondly, the passive resistance behavior of a human MCP joint is analyzed during a full cycle of flexing and moving back to the initial position. Thirdly, the tool is used for an inter-day evaluation to determine the precision of the tool when used on human MCP joints (expressed as coefficient of variation (CV)). Finally, a repetition test is performed to determine the influence of setup-related causes on the precision of the tool.

Results
The linear spring evaluation showed that the mechanical system of the tool is very precise, exhibiting a coefficient of variation of 0.6%. The continuous cycle evaluation showed that the passive resistance of the MCP joint behaves nonlinearly and that the MCP joints have viscoelastic properties. The calculated CV for the inter-day precision evaluation is equal to 33.0%. The CVs for the repetition tests were equal to 13.4% and 27.7% for the fixed and non-fixed approaches respectively. Other evaluated aspects are the weight of the tool (1.6 kg), volume of the tool (10.8 dm³), comfort for the user (9 out of 10), and full six-joint assessment time (7.5 min).

Discussion
The CV of 33.0% for the inter-day precision evaluation is too large to subject the tool to RA patients. However, the repetition test showed when the hand under assessment is consistently aligned and orientated, the system is much more precise, exhibiting a CV of 13.4%. So, when design components are included that assist with the alignment and orientation of the hand under assessment, the tool can be subjected to actual RA patients to evaluate the increase in passive MCP joint stiffness during (the early stages of) inflammation.