Background/Purpose

A variety of pain and function instruments are often measured in osteoarthritis (OA) clinical trials. Instruments with maximal sensitivity to change are preferred as the primary measure in trials. Shown to be sensitive to change in rheumatoid arthritis, patient preference instruments, in which a patient nominates a painful activity to track, have not been tested in OA trials. Ideally, to compare the sensitivity to change of outcome measures, trials with a positive result need to be studied. We compared the sensitivity to change of pain and function self-report outcomes in 2 OA trials that reported statistically significant improvements in pain/functional status outcomes.

Methods

We used data from a controlled trial of a knee brace (BRACE; n = 126) and an uncontrolled trial of intraarticular steroid injection (TASK; n = 120). For both trials, eligible subjects had to meet ACR criteria for knee OA and have significant knee pain. In addition to completing KOOS and WOMAC questionnaires, and questions about overall knee pain, subjects were asked to nominate an activity that commonly caused them knee pain, and to rate pain during that activity throughout the trial. Standardised response means (SRMs) were generated for multiple time points in both trials, in the active treatment group alone, and comparing treatment to control, where possible. Additionally, we tested correlations between the study outcomes and an anchor variable: a 5-point likert scale of patient-perceived change in pain, using ordinal regression to determine the odds of an improvement in pain for a 1 SD increase in each pain outcome.

Results

Both trials reported a positive effect of the intervention on pain and function. Pain on nominated activity visual analogue score (VAS) produced SRMs that were at least as high as other pain outcomes, and usually higher (table 1). Sensitivity to change of the other outcomes was less consistent. Patient perceived improvement in pain was more strongly associated with pain on nominated activity than the other pain or function outcomes (table 2).

Conclusion

Pain on nominated activity may be an extremely sensitive outcome for OA trials, and appears more strongly associated with perceived pain improvement than other currently used outcome measures.

Table 1: Standardised Response Means (SRMs) for the TASK and BRACE Trials.

 

 

Table 2: Association between Different Pain Outcomes and Patient Perceived Improvement in Pain. Odds Ratios Indicate Odds of Improvement in Patient Perceived Pain, for a 1 SD Improvement in the Outcomes Listed.

 

Outcome	BRACE Study – 6 week change, all persons pooled	BRACE Study – 12 week change, all persons pooled	TASK Study – Change at 1 week follow-up
Pain on nominated activity VAS	2.09	2.57	6.12
Pain in last week VAS	1.94	2.15	4.45
Global pain VAS	–	–	2.36
KOOS pain subscale	1.63	1.79	1.87
KOOS symptom subscale	1.64	1.62	1.88
KOOS activities of daily living subscale	1.67	1.70	1.86
WOMAC pain subscale	1.60	1.71	1.84
WOMAC stiffness subscale	1.57	1.56	1.79
WOMAC function subscale	1.67	1.70	1.86

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/sensitivity-to-change-of-patient-preference-outcome-measures-for-pain-in-trials-of-patients-with-knee-osteoarthritis/