Background/Purpose

Clinically important worsening in OA has not been well defined. Validated worsening criteria are important for research but also clinical practice to make informed treatment choices.  The goals of this study are 1) to select candidate clinical worsening criteria and 2) to validate criteria for clinically important worsening.

Methods

Data were used from a cohort of knee and hip OA outpatients visiting our department who received standardised evidence-based tailored conservative treatment in a stepped-care format for 3 months. The development cohort comprised 218 patients with three-months follow up and the validation cohort consisted of 296 patients with two-years follow up. For this study baseline and three month data were used.

In the first round, an expert group (methodologists, orthopaedic surgeon, physical therapists, psychologist , rheumatologists) reviewed previously proposed criteria for clinical worsening and selected on the basis of consensus and face-validity the criteria that should be tested. Furthermore,  the expert group decided that newly defined criteria of worsening should contain both pain and function and optional PGA and embed both an absolute and a relative change. Then, 15 new criteria of worsening criteria were constructed. In the second round, first we examined the sensitivity and specificity of the newly developed criteria of worsening in development cohort using an anchor-based 7-point Likert transition scale (much worse-much better). Second, we evaluated all the selected criteria from the literature and newly developed criteria of worsening criteria in the validation cohort using the transition scale.

Results

The expert group’s review of existing criteria yielded five criteria that were included in the validation round and the sensitivity of these previously proposed criteria for clinical worsening was low while specificity was high. Of the expert group’s proposed criteria, the sensitivity and specificity of the newly developed worsening criteria ranged from 1 to 65% and 60 to 97% respectively. The three newly defined sets of criteria that performed best incorporated smaller relative and absolute changes compared with improvement criteria. Set 1 and 3 performed best, with sensitivity and specificity ranging between 50 and 75% in both the development and the validation cohort (Table 1).

Conclusion

Previously proposed criteria for clinical worsening in knee or hip OA are specific but lack sensitivity. Our results suggest that, compared with improvement criteria, criteria for worsening should incorporate relatively small absolute and relative changes. Our newly developed criteria pairs acceptable sensitivity with acceptable specificity and can be used to measure clinically important worsening in knee/hip OA studies and clinical care.  

Table 1. Sensitivity and specificity of newly developed worsening criteria, in the development and validation cohort